ROAD ECOLOGY OF A NORTHERN POPULATION OF BADGERS (TAXIDEA TAXUS)
IN BRITISH COLUMBIA, CANADA
BY

RICHARD WALTER KLAFKI
BSC, UNIVERSITY OF NORTHERN BRITISH COLUMBIA, 2001

A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE
IN
ENVIRONMENTAL SCIENCE
THOMPSON RIVERS UNIVERSITY
KAMLOOPS, BRITISH COLUMBIA, CANADA
JUNE 2014

Thesis supervising committee:
Karl Larsen (PhD), Thesis Supervisor and Professor in Natural Resource Sciences,
Thompson Rivers University
Tom Dickinson (PhD), Dean of Science and Associate Professor in Biological Sciences,
Thompson Rivers University
Richard Brewster (PhD), Professor in Mathematics and Statistics,
Thompson Rivers University
Roger Packham, R.P. Bio., Ministry of Environment (Retired), 100 Mile House, BC
Chris Johnson (PhD), External Examiner, Associate Professor in
Ecosystem Science and Management, University of Northern British Columbia

“The last word in ignorance is the man who says of an animal or plant: 'What good is it?”
― Aldo Leopold

ii

ACKNOWLEDGMENTS
I would like to acknowledge the support, advice, and most importantly the patience of
my supervisor, Dr. Karl Larsen at Thompson Rivers University. His dedication to
conservation and research has provided me the opportunity to conduct research on this
intriguing and challenging animal and I thank him for the experience. Instrumental in the
initiation of the project, Roger Packham provided his expertise and knowledge of the region
with me and provided a supportive and enthusiastic work environment. I would also like to
thank the British Columbia Ministry of Transportation and Infrastructure for providing
financial support for the project. Special thanks to Dr. Helen Schwantje and Eric Lofroth for
their support. Thanks also to my committee members Dr. Tom Dickinson and Dr. Rick
Brewster for their feedback and patience and Dr. Chris Johnson from UNBC who agreed to be
my external reviewer on short notice.
A huge thanks to all the members of the Badger Recovery Team for advice and
logistical support. Thanks also to all the veterinarians and techs that assisted in the project
with a special thanks to the mobile vet unit of Dr. Malcolm McAdie. Thanks to Paul Williams
and staff of the BC Wildlife Park and ‘Bloom’ the badger, for providing me with an excellent
candidate to test out the GPS harness system. Professional piloting from Lawrence Aviation
for aerial telemetry was much appreciated. Outstanding data collection in the field was carried
out by Hamish Nicolson, Nikki Heim, Jonquil Crosby, Ashley Burkatsky, Julie Kline, Mike
Packham, and members of the Canoe Creek Band. I would also like to thank my fellow
students and friends for support and making life interesting around campus. Thanks to all the
people who offered me hospitality during the field seasons, particularly the Bloms’, Witt’s
End Guest Ranch, and the Packham family. A special thanks to all the landowners in the south
Cariboo (over 64) that provided access to their properties and allow badgers the room to roam!
Finally, a heartfelt thank you to my family and close friends for their continual encouragement
and support during my academic pursuits!
The research was also supported by the Natural Sciences and Engineering Research
Council of Canada, Habitat Conservation Trust Foundation, BC Forest Science Program, BC
Conservation Foundation, BC Ministry of Environment. This project was approved by
Thompson Rivers University Animal Care Protocol no. TRU2007-8, and BC Ministry of
Environment Permit no. WL07-36484.

iii

ABSTRACT

The badger (Taxidea taxus) is one of British Columbia’s (BC) most endangered mammals,
with estimates placing the current provincial population at fewer than 340 adults. Previous
research in the province has isolated two main factors contributing to the decline of badgers
in BC: (1) a historical and current decline in habitat, and (2) road mortality caused by
highways running through badger habitat. This intensive badger radio-telemetry study was
initiated in 2007 to document badger ecology and movements in order to investigate why
badgers are susceptible to vehicle collisions and possible mitigation methods.
From 2007-2009, a subset (11♀, 5♂) of the badger population in south-central BC were
captured and outfitted with VHF radio-tags to track their location and movements. In
addition, the systematic collection of hair to identify individuals using DNA was employed to
assist in tracking movements, dispersal, and aid in identifying carcasses. Locations for study
animals were used to determine fixed kernels (95% FK) for home range analysis. Mean home
ranges for males were 253.1 km2 while females had much smaller home ranges of 29.4 km2.
In addition to their extremely large home ranges, males displayed extensive movements
during summer, while reproductive females maintained the smallest home ranges and
restricted their movements during the kit rearing period. Movements of adult females with
kits were influenced by time of kit emergence from the den, litter size, and period of time
before natal den abandonment.
Newly emerging GPS technology was used to reveal extensive movements during summer
that conventional radio-telemetry had underestimated. Movement rates were five times
greater for GPS outfitted badgers versus conventional radio-telemetry estimates [GPS 4,801
m/day (n = 5): VHF radio 840 m/day (n = 16)] suggesting that badger movements and road
crossings were underestimated by previous badger ecology studies.
In an attempt to document the frequency and ratio of above- and below-grade road crossings,
remote motion cameras were placed at existing underpass structures to compare telemetry
locations and GPS movements with camera data. Both marked and unmarked badgers were
recorded using a variety of structures. Data revealed that both genders and family groups
used culverts and livestock underpasses to pass under roads, with 500 mm culverts being

iv

used most frequently. Both above- and below-grade road crossings were detected, with
frequency varying by individual, time of day, and seasonality.
Animals whose home ranges bisected highways were repeatedly observed crossing
highways, with half of the study animals succumbing to collisions with vehicles (8 of 16).
Due to their large home ranges, all five males in the study eventually crossed highway
corridors with four of five males dying on these major roadways. Sites of badger road
mortalities were spatially defined to identify areas of concentrated road mortality sites or
‘hotspots’ on major roadways in south-central BC.
Despite the high percentage of study animal road mortality, DNA data and population
modeling suggests an increasing population of badgers in the regional study area. At the
present time, the high incidence of road mortality in the study area seems to be negated to
some extent by strong recruitment, hence the estimate of increasing population size.
Home range size and movement rate estimates for badgers in this study are some of the
largest reported in BC. These phenomena may be explained by the patchy distribution of
resources near the northern range limit of badgers in BC, such as suitable soils for burrowing
and distribution of prey species. These limited resources often coincide with human
populated areas where development of transportation corridors and agriculture often occur.
These human-induced pressures on badger habitat often conflict with seasonal movement
patterns of badgers (i.e., breeding season) leading to habitat fragmentation and direct
mortality. The ability to increase our understanding of these movement patterns and provide
mitigation measures (e.g., road underpasses, land conservation) that reduce direct mortality
and increase productivity should allow for the continued existence of this species near their
range limit in south-central BC, Canada.
In conclusion, I present a focused discussion on how an understanding of the ecology of
badgers (particularly movements) may be integrated into habitat protection and restoration
plans, including mitigation for current and future road infrastructure projects, thereby
reducing impacts on this species.

v

TABLE OF CONTENTS
ACKNOWLEDGMENTS ........................................................................................................ ii
ABSTRACT ............................................................................................................................. iii
TABLE OF CONTENTS .......................................................................................................... v
LIST OF FIGURES ................................................................................................................ vii
LIST OF TABLES ................................................................................................................... ix
CHAPTER 1 – MESOCARNIVORE BIOLOGY AND CONSERVATION: THE BADGER
(TAXIDEA TAXUS) IN BRITISH COLUMBIA, CANADA ............................................. 1
The mesocarnivore ................................................................................................................ 1
The North American badger ................................................................................................. 2
The badger as a North American mesocarnivore .............................................................. 2
Badgers in British Columbia, Canada ............................................................................... 4
Thesis goals and structure ..................................................................................................... 9
Study site description .......................................................................................................... 10
Literature cited .................................................................................................................... 16
CHAPTER 2 – ROAD ECOLOGY OF BADGERS (TAXIDEA TAXUS) IN SOUTHCENTRAL BRITISH COLUMBIA, CANADA ............................................................. 21
Introduction ......................................................................................................................... 21
Study area............................................................................................................................ 24
Methods............................................................................................................................... 26
Live-capture and radio-telemetry .................................................................................... 26
Reproduction ................................................................................................................... 33
Mortality ......................................................................................................................... 33
Distribution and abundance ............................................................................................ 35
Highway structures influencing movements and mortality ............................................ 40
Results ................................................................................................................................. 41
Live-capture and radio-telemetry .................................................................................... 42
Home range ..................................................................................................................... 42

vi

Dispersal ......................................................................................................................... 49
Reproduction ................................................................................................................... 49
Mortality ......................................................................................................................... 52
Population modeling ....................................................................................................... 57
Highway structures influencing movements and mortality ............................................ 60
Is the population significantly impacted by road mortality? .......................................... 68
Mitigation and/or compensation ..................................................................................... 70
Conclusion ...................................................................................................................... 74
Literature cited .................................................................................................................... 75
CHAPTER 3 – REDUCE MORTALITY AND INCREASE RECRUITMENT; A
STRATEGY FOR MAINTAINING BADGERS (TAXIDEA TAXUS) IN SOUTHCENTRAL BRITISH COLUMBIA ................................................................................. 81
Introduction ......................................................................................................................... 81
Overview of study findings ................................................................................................. 81
Management and mitigation ............................................................................................... 83
Habitat protection / restoration (increase recruitment): .................................................. 83
Mortality reduction (highway specific): ......................................................................... 84
Future study .................................................................................................................... 90
Discussion ........................................................................................................................... 91
Literature cited .................................................................................................................... 93
Appendix A. Reproductive parameters for female badger natal dens monitored during
the study in south-central BC, Canada 2008-2010. ..................................................... 95
Appendix B. Fates of radio-tagged badger study animals, capture dates, known mortality
dates, fate, and number of days radio-monitored as of June 2010. ............................. 96
Appendix C. Poster for solicitations for badger sightings from the public. ........................ 97
Appendix D. Public brochure handout (front and back) distributed in the south Cariboo
region, 2007-2010. ....................................................................................................... 98

vii

LIST OF FIGURES
Figure 1.1. North American distribution of four subspecies of badger (Taxidea taxus) (from
Newhouse and Kinley 2001). ............................................................................................ 5
Figure 1.2. Map of badger (Taxidea taxus) distribution in British Columbia, Canada (Weir
and Almuedo 2010). .......................................................................................................... 7
Figure 1.3. Regional badger study area (outer) and local study area (inner) with highway
locations near the community of 100 Mile House (N51° 39΄ 11˝, W121° 17΄ 29˝) in
south-central British Columbia, Canada. ........................................................................ 11
Figures 1.4A and 1.4B. Aerial photo of Highway 97 bisecting valley bottom habitat in
northern portion of study area (above); and rolling hills with grassland/pasture habitat in
central portion of study area (below) in south-central BC, Canada. ............................... 13
Figures 1.5A and 1.5B. Examples of badger habitat adjacent to wetlands. Field technician
documenting a burrow in fine-textured soils upslope of a wetland, with grassland
transitioning to Douglas fir dominated forests in background (top); Typical badger
habitat adjacent to wetlands in the central portion of study area (below). ...................... 15
Figure 2.1. Regional study area (outer) and local study area (inner) boundaries with highway
locations near the community of 100 Mile House (N51° 39΄ 11˝, W121° 17΄ 29˝) in
south-central British Columbia, Canada. ........................................................................ 25
Figures 2.2A and 2.2B. Harness system testing on tame, captive badger illustrating sternum
strap for stabilization of GPS datalogger backpack (top), and removing the harness
system (bottom) at British Columbia Wildlife Park, Kamloops BC, Canada, June 2008.
......................................................................................................................................... 29
Figure 2.3. Dorsal view of a wild free-ranging male badger outfitted with GPS datalogger
nine days after deployment exiting from a burrow near Meadow Lake, BC, 14 August
2009. Photo taken by remote motion camera. ................................................................. 30
Figure 2.4. Remote infrared motion camera housed in security box attached to stand
deployed at badger burrow. ............................................................................................. 34
Figure 2.5. Schematic diagram (l) of hair-snagging device deployed in badger burrow and
close up photo of hair-snagger (r) with suitable hair sample for DNA analysis. Each of
the two snagging pinned knaplock segments were approximately 2.5 cm x 4 cm in
dimension. ....................................................................................................................... 36
Figure 2.6. Hair-snagger deployed in active badger burrow revealing suitable hair sample for
DNA analysis. Photo taken near 100 Mile House, BC by P. Verkerk. ........................... 36
Figure 2.7. Adult male badger fixed kernel (FK 95%) polygons and 100 minimum convex
polygons (100% MCP, straight lines) home ranges (n=5) in south-central BC, Canada,
2007 – 2010. .................................................................................................................... 44
Figure 2.8. Female badger home ranges (n = 11) 95% FK polygons and 100% MCP (straight
lines) in south-central BC, Canada, 2007 – 2010. ........................................................... 46
Figure 2.9. Detailed GPS track of adult female badger's (BF05) movements crossing
Highway 24 near Bridge Lake, BC, 04 - 17 August 2009. Squares represent sequential

viii

GPS locations acquired every 15 minutes. Inset map: Entire GPS track overview and
home range outline (95% FK) derived from VHF radio-telemetry................................. 50
Figure 2.10. Kaplan-Meier survival curves for male and female radio-tagged badgers that
died on roadways in south-central BC, 2007-2009. ........................................................ 53
Figure 2.11. Percentage of badger-vehicle collisions in relation to average daily traffic
volume (ADT)* by month on Highway 97 in south-central British Columbia, Canada,
2004 – 2010. *Source: Ministry of Transportation and Infrastructure, Traffic Data
Program, 2004 – 2008. .................................................................................................... 55
Figure 2.12. Plotted road mortality locations with 70% fixed kernel utilization distribution
contours revealing three ‘roadkill hotspots’ in relation to major roads in the northern
portion of the study area in south-central BC, 2004 – 2010. Sixty percent of badger road
mortality in the region occurred in these three zones along Highway 97 in the study area
from 2004 – 2010. ........................................................................................................... 56
Figure 2.13. POPAN model derived population estimate of badgers across the regional study
area in south-central British Columbia, Canada, 2003-2008. The 2003 estimate is based
on minimum number of badgers detected alive. Bars represent +/- 1 standard error. .... 61
Figure 2.14. Number of underpass crossing structures monitored (n = 41) by size and
percentage of use by badgers in south-central BC, 2008 – 2009. ................................... 63
Figure 2.15. Number of underpass crossings documented by identified study animals, known
juveniles, and unknown badgers in south-central BC, 2008 – 2009. .............................. 63
Figure 2.16. Adult female (BF01) with male kit (right) exiting 500 mm culvert under
Highway 97 in south-central BC, 21 July 2009. ............................................................. 65
Figure 2.17. Facial pattern recognition was used to identify unique individual badgers that
passed through underpass structures in south-central BC, 2008 – 2009. Radio-tagged
female (left) and unknown male (right). ......................................................................... 65
Figure 2.18. Adult female badger (BF01) exiting culvert with GPS datalogger backpack in
south-central BC, 01 August 2008. ................................................................................. 73
Figure 3.1. Recently installed badger-specific underpass crossing structure along Highway 97
south of 100 Mile House BC, Canada, 2010. .................................................................. 86
Figure 3.2. Suitable existing underpass structure (500 mm corrugated pipe) with earth
substrate bottom used by badgers to pass under Highway 97 north of 100 Mile House
BC, Canada, 19 August 2008. ......................................................................................... 87
Figures 3.3A and B. Examples of raised banks adjacent to highway corridor. Badger road
mortality site (top) where no crossing structures were present nearby (> 250 m) and
successful crossing site used repeatedly by badger (below). Both sites along Highway 97
north of 100 Mile House BC, Canada. ............................................................................ 89

ix

LIST OF TABLES
Table 2.1. Annual and seasonal* 95% fixed kernel home ranges (95% FK) for all study
animals, July 30, 2007 - January 31, 2010. Sample sizes for each seasonal estimate are
provided in the 5 rightmost columns. Seasons are defined as follows: Spring (MarchMay), Summer (June-August), Fall (September- November), and Winter (DecemberFebruary). ........................................................................................................................ 43
Table 2.2. Average movement rates (m/day) derived from conventional radio-tagged badger
data monitored between 2007 and 2010 for all seasons in south-central BC.................. 48
Table 2.3. Summary of GPS datalogging deployment on four badgers in south-central BC,
July - August 2008 and 2009. .......................................................................................... 48
Table 2.4. Minimum detected dispersal distances of badger kits as determined by DNA or
live-capture mark/recapture at burrows. Detection method describes whether the animal
was live-captured or DNA hair-snagged at burrow......................................................... 51
Table 2.5. Suite of top-ranked candidate models derived by the Reduced Burnham (Both)
Jolly-Seber method. These models provides estimates for survival (S), recapture (p), and
recovery (r) of marked individuals based on corrected Quasi-Akaike Information
Criterion (QAICc) with an adjusted ĉ=1.174. Model names indicate if survival (S) or
probability of capture (p), recovery (r), and fidelity (F) were held constant (.), or varied
with group (g) or time (t). ∆QAICc is the difference in QAICc values between a model
and the most parsimonious model. ‘Model weights’ indicates the support for a model.
Number of parameters specifies the number of variables estimated for a model.
QDeviance is the deviance of a model from the full general model. .............................. 58
Table 2.6. Output from program MARK vers. 6.1 for most parsimonious model providing
population parameter estimates for survival by year (S), probability of recapture (p), and
probability of recovery of dead marked individuals (r) for the regional badger population
in south-central BC, Canada, 2003 – 2008. ..................................................................... 59
Table 2.7. Pradel-λ badger population model selection based on Akaike Information Criterion
(AIC). Model names indicate if population growth (lamda, λ) was held constant (.) or
varied by time (t). Model weight indicates the support for a model. Number of
parameters specifies the number of variables estimated for a model. Pradel‐λ model does
not allow calculation of weighted averages for Φ and p, and therefore it does not provide
estimated values for them. ............................................................................................... 62
Table 2.8. Badger abundance estimates derived from both the POPAN model and estimates
of λ derived from Pradels-λ model. A similar trend in population growth of badgers in
the regional study area from 2003-2008 is shown using both methods. ......................... 62

1

CHAPTER 1 – MESOCARNIVORE BIOLOGY AND CONSERVATION: THE
BADGER (TAXIDEA TAXUS) IN BRITISH COLUMBIA, CANADA
THE MESOCARNIVORE
A combination of rarity, anthropogenic impacts, and political factors has put roughly
a quarter of all terrestrial mammal species at risk of extinction (Ceballos et al. 2005). The
underlying cause of virtually all recent and ongoing declines of these animals is the growth
of human populations and associated impacts such as habitat loss, over-exploitation, and the
spread of invasive species (Cardillo et al. 2004). Nowhere has this decline been more
precipitous than within the Order Carnivora.
Carnivores, particularly the larger species, are ecologically important because the loss
or gain of a few individuals can cause strong predator-driven effects or indirect behavioural
effects (e.g., fear-driven foraging) on communities and ecosystems (Ripple and Beschta
2004a, Ray 2005, Roemer et al. 2009). The loss of large carnivores also likely carries
broader implications for the maintenance of biodiversity as a result of indirect effects at
lower trophic levels (Crooks and Soule 1999, Henke and Bryant 1999). This phenomenon
has been termed a ‘trophic cascade’, namely a predator/prey interaction that has effects
throughout more than one level in a given food web (Hairston et al. 1960, Paine 1980).
Although the classic trophic cascade is based on a three-tiered system consisting of predators,
herbivores, and plants, cascades can in theory involve more than three trophic levels and may
apply to any multilink linear food web interaction (Hairston et al. 1960, Polis et al. 2000).
The importance of large carnivores notwithstanding, most mammalian predators are
neither large, nor at the apex of trophic food webs. A case in point is the so-called
‘mesocarnivores’ that often are limited by effects from higher up in the food web (i.e. “topdown effects”). Mesocarnivores may therefore have their ecological impact on the broader
community mediated by the presence of other larger predators (Gittleman and Gompper
2005). But, given their smaller size and ability to thrive in diverse habitats, mesocarnivores
are usually more abundant than large carnivores, yet their impact within communities has
been generally assumed to be relatively minor (Roemer et al. 2009). Further, this group
represents an ecologically diverse and influential guild of vertebrates that are capable of
structuring the ecological community in which they occur (Buskirk and Zielinski 2003).

2

Recent arguments by Roemer et al. (2009) suggest the importance of mammalian
mesocarnivores in ecosystems may be understated. Using theory and empirical data, they
point out that the impact of these animals may be especially profound under three scenarios:
(1) when larger carnivores are absent in an ecosystem, (2) on island or mainland ecosystems
with relatively simple ecological communities, and (3) where they represent non-native
introductions [e.g., foxes (Vulpes vulpes) in Australia (Banks 1999)]. All told,
mesocarnivores may be fundamentally important drivers of ecosystem function, structure, or
dynamics (Boitani 2001, Meserve et al. 2003, Donadio and Buskirk 2006, Maron et al. 2006,
Jones et al. 2008).
Most often it is the large carnivores, such as wolves (Canis lupus), grizzly bears
(Ursus arctos), and mountain lions (Puma concolor) that are considered the most useful
species for conservation planning (Carroll et al. 2001). More recently, conservation biologists
have recognized that mesocarnivores can have the same role, not only because the home
range sizes of some species [e.g. Wolverine (Gulo gulo), lynx (Lynx canadensis)] are similar
to or greater than those of large carnivores, but because the extirpation of large carnivores in
many regions has promoted mesocarnivores to the role of top predators (Buskirk and
Zielinski 2003). Including a combination of focal carnivore species, especially
mesocarnivores that specialize on habitats of concern, may ultimately be the best approach to
large-scale monitoring and conservation of natural landscapes and ecosystem processes
(Noss et al. 1996).
THE NORTH AMERICAN BADGER
The badger as a North American mesocarnivore
Most ecological research on badgers (Taxidea taxus) across North America has been
limited to the western United States in the 1970s and 1980s (Sargeant and Warner 1972,
Lindzey 1978, Messick and Hornocker 1981, Lampe 1982). Recently however, projects have
been focused across their range where populations may be declining due to environmental
and anthropogenic factors (Warner and VerSteeg 1995, Weir et al. 2003, Paulson 2007,
Duquette 2008, Kinley and Newhouse 2008, Quinn 2008, Hoodicoff et al. 2009).
The North American badger is one of the largest members of the Family Mustelidae
(the ‘weasel family’). A fossorial mesocarnivore, they are adapted for digging to obtain food

3

and shelter. The well-developed forelimbs and claws are ideal for burrowing with a conical
and wedge-shaped head for thrusting into small animal burrows. Adult badgers usually range
in length from 72 to 91 cm, including the short bushy triangular tail, which is typically 10 to
14 cm long. Adult females weigh between 6 and 9 kg, while larger adult males weigh
between 9 and 13 kg.
Many studies across the range of the badger in North America have been conducted
and the ecological characteristics for the species have been described. The home ranges of
both male and female badgers expand during the breeding season, indicating that both
genders travel more extensively at that time to find mates (Sargeant and Warner 1972,
Messick and Hornocker 1981). Males have larger home ranges and are likely to overlap with
the home ranges of several females (Lindzey 1978, Minta 1993). Badgers generally mate in
late summer between July and August. Young are born blind, furred, and helpless sometime
in late March to early April. Their eyes open at four to six weeks of age. Weaning occurs
around mid-May when the young are about half-grown. Juveniles are then taught to hunt
around mid-June to early July when they are about two-thirds grown. The number of young
reportedly ranges from one to seven; two to three young are most common (Kinley and
Newhouse 2008). Badgers are polygamous forming loose pair bonds in late summer. They
are territorial, particularly the females, and males likely fight over a female (Long and
Killingley 1983). Dispersal and breakup of family units usually occurs in autumn when the
female stops bringing food to the young badgers (Sargeant and Warner 1972, Messick and
Hornocker 1981).
Activity patterns of badgers vary not only by time of day but are also influenced by
seasons. Badgers are most active in summer, and least in winter. They are also most active at
night, but reports of badgers being active in the day are common. They are not true
hibernators, but spend much of the winter in cycles of torpor characterized by a reduction in
heart rate and body temperature (Long and Killingley 1983). Badgers will accumulate stores
of fat in the autumn and early winter when much of their prey hibernates and are thus easier
to catch.
Badgers use multiple burrows within their home range and are capable of digging
several in a single night. They will frequently reuse burrows and multiple badgers may use

4

the same burrow in an area over several generations. While usually considered solitary
animals, sociality among badgers has been observed, particularly family units in the summer
(Messick and Hornocker 1981).
Badger diets in North America can vary depending on the geographic range where
they occur. They have been described as an opportunistic predator, preying on various other
species (Lampe 1982, Messick 1987, Sovada et al. 1999, Hoodicoff 2003, Michener 2004,
Kinley and Newhouse 2008). Small mammals form the majority of their diet with an
emphasis on fossorial colonial rodents such as ground squirrels (Spermophilus spp.) and
marmots (Marmota spp.), species on which badgers have become specialists in exploiting as
food resources (Michener 2004, Kinley and Newhouse 2008). Other common mammalian
prey species include northern pocket gopher (Thomomys talpoides), muskrat (Ondatra
zibethicus), leporids (rabbits and hares), and various microtine rodents (e.g., voles)
(Newhouse and Kinley 2001, Hoodicoff 2003). Less common but other reported prey has
included insects, birds (e.g., waterfowl, ground nesting), reptiles, amphibians, and fish.
Badgers are considered an important ecological driver for grassland and open forest
ecosystems as they not only influence rodent prey populations but also geophysical processes
(Eldridge and Whitford 2009). Both foraging and resting burrows dug by badgers form
important habitat elements for other species that use vacant burrows for nesting sites or
thermal cover, such as burrowing owls (Speotyto cunicularia) and several snake species (e.g.,
western rattlesnake, Crotalus oreganus) . Badger excavations and associated soil mounds are
important features that alter the composition of grassland and open forest ecosystems by
influencing a wide range of processes including water infiltration, soil aeration, organic
decomposition rates, vascular plant diversity, and support for a variety of soil invertebrates
(Eldridge 2004). Thus, this mesocarnivore shapes and influences specific habitats across its
geographic range, and its presence is an excellent indicator for identifying properly
functioning grassland / open forest ecosystems.
Badgers in British Columbia, Canada
Badgers (Taxidea taxus) occur at their northern range limit across southern Canada
(Figure 1.1). Of the four subspecies of badgers occurring in North America, only T.t.
jeffersonii occurs in British Columbia (BC) with an estimated population of less than 340

5

Figure 1.1. North American distribution of four subspecies of badger (Taxidea taxus)
(from Newhouse and Kinley 2001).

6

adult individuals. The Rocky Mountains create the boundary between the jeffersonii
subspecies of badgers (west of the Rockies) and T. t. taxus in the prairie provinces. Further
south, the two races become intergraded on the Great Plains.
In BC, the range of the jeffersonii badger extends along the dry southern interior
valleys, occurring west to the Coast Mountains, north to the Fraser Plateau and east to the
Alberta border along the southern Rocky Mountain Trench (Figure 1.2).
Two major habitat features considered vital to badgers are suitable densities of prey
and appropriate substrates in which to dig burrows (Rahme et al. 1995). Studies in BC
supports this assumption, with the animals commonly being associated with grasslands and
open forests that provides these resources (Apps et al. 2002, Weir et al. 2003).
Badgers are frequently found in low-elevation human-altered landscapes, such as
gravel pits, golf courses, agricultural fields, forest-harvesting blocks, along highways, and
ranchlands where disturbed soils offer ease of digging and rodent prey is plentiful. These
habitats are also associated with linear features, such as roads and trails that offer the shortlegged badger convenient travel corridors. Very few badger observations occur in the interior
wetbelt with the Columbia Mountains posing as a substantial barrier to movement and
dispersal. Thus, the biogeography of BC restricts badger distribution into disjunct regional
sub-populations where maintaining genetic diversity becomes of critical importance in the
face of declining populations across their range.
The conservation status and ecology of badgers in BC came to light during the late
1990s, when a sharp decline in their populations became apparent (Newhouse and Kinley
1999). While there is very little reliable historical data, it has been estimated that provincial
populations have likely declined since the 1920s when the number of badgers trapped
annually were greater than the current population estimate of 230-340 animals (jeffersoniiBadger-Recovery-Team 2008). While the global status of the North American badger is
considered secure (G5, NatureServe), in BC the jeffersonii subspecies is considered
threatened by the BC Conservation Data Centre, and is federally ranked as ‘endangered’ by
the Committee on the Status of Endangered Wildlife in Canada (COSEWIC).
Recent studies in BC have identified declining numbers of badgers in regional
subpopulations of the Thompson-Okanagan and East Kootenay, including recent extirpation

7

Figure 1.2. Map of badger (Taxidea taxus) distribution in British Columbia, Canada (Weir and Almuedo 2010).

8

from historical range in the Upper Columbia valley in the East Kootenay region (Weir et al.
2003, Kinley and Newhouse 2008, Hoodicoff et al. 2009). These finding in conjunction with
genetic work conducted by Kyle et al. (2004) suggest that badger populations in BC occur in
units that may have independent demographics. The primary factors limiting badger
populations in BC are believed to be related to road mortality, habitat impacts, and prey
constraints (Newhouse and Kinley 2001, Weir et al. 2003), although the exact effects of
modification, alienation, and loss of habitat on badger populations in the province are largely
unknown (jeffersonii-Badger-Recovery-Team 2008).
Badgers in south-central BC (the ‘Cariboo’ region) have only recently been studied,
in part because of the overall growing concern for badgers in the province. Prior to 2003,
little was known about these animals other than the occasional sighting and vehicle collision
mortality report. Fortunately, the fact that pressure on the population from human activities
may be relatively recent suggests there may still be time to prevent the drastic decline of the
animals seen in other regions. In 2003, the BC Ministry of Environment launched a project
aimed at identifying the distribution and abundance of badgers in the region through burrow
searches and solicitation of sightings of badgers and their burrows (Hoodicoff and Packham
2007). The project’s objectives later evolved to include developing and testing non-invasive
DNA hair-snagging techniques for estimating badger demographics in the region. A number
of recovery activities in the Cariboo region resulted from this research, including refining
hair-snagging protocols and creation of managed Wildlife Habitat Areas to maintain and
improve habitat for badgers on public lands. In response to road mortality recorded during
the project, badger road crossing signs to warn motorists were erected along Highway 97 in
the central Interior.
It is well known that some carnivores have been severely affected by the expansion of
transportation infrastructure (Ferreras et al. 1992, Proctor et al. 2005, Taylor 2005), however
little is known about the response on populations due to mortality (Noss et al. 1996, Blanco
et al. 2005). Roads can impede the movement of animals between resource patches,
subdivide populations, and increase the risk of road mortality due to collisions (Forman et al.
2003). The leading cause of mortality for badgers in BC is badger-vehicle collisions on major
highways (Weir et al. 2003, Kinley and Newhouse 2008). This pattern is not restricted to BC,
as American wildlife biologists also have reported elevated mortality due to collisions with

9

vehicles (Messick 1987), and the impacts of vehicles on the Eurasian badger (Meles meles)
are well known (Clarke et al. 1998).
Badgers are vulnerable to highway mortality for several reasons: (1) highways are
constructed in valley bottom habitats preferred by badgers, (2) badgers’ large home ranges
and peak movement rates (mating period) coincide with peak summer traffic volumes; (3)
roadside right-of-ways are attractive to badger prey, and (4) badgers are most active at night,
when drivers have greatest difficulty seeing smaller animals on the highway. In BC,
proportion of mortality attributed to badger-vehicle collisions ranged from 13% in the East
Kootenay to an astounding 86% in the Thompson region (Weir et al. 2003, Kinley and
Newhouse 2008). Further habitat loss and degradation will likely cause further increases in
road mortality.
Understanding the anthropogenic impacts of roads on badgers in the region is
particularly pressing given that this population occurs at the extreme northern edge of their
range. Peripheral populations are presumably encountering natural, limiting factors (e.g.,
soils, climatic factors), so the additive or synergistic effects of a factor like road mortality
may be accentuated and lead to local extirpation (Kirkpatrick and Barton 1997, Fraser 2000,
Vucetich and Waite 2003). Further, peripheral populations are likely to experience different
regimes of natural selection than central populations, as geographically outlying populations
are likely to occur in ecologically marginal or stressful conditions (Lesica and Allendorf
1995). Many species at the edge of their range occur in unusual or atypical habitats (van
Woudenberg and Christie 1997). Thus, peripheral populations are expected to be genetically
distinct because of divergent natural selection (Lesica and Allendorf 1995). This theory is
supported by genetic sampling of badgers in BC, as Kyle et al. (2004) found genetically
distinct units between populations separated by mountain ranges. Therefore, badgers in
south-central BC have a high conservation value as distinct traits found in this population
may be crucial for allowing the species to adapt in the face of environmental change.
THESIS GOALS AND STRUCTURE
As mentioned above, the monitoring of badgers in the Cariboo region of BC was
initiated in 2003 by government biologists to determine burrow distribution and identify
individuals. Using newly emerging DNA hair-snagging methods in 2007, I expanded the

10

scope of the Cariboo region badger project to include live-capture and VHF radiomonitoring, in tandem with newly emerging micro-GPS technology. The overarching
objective of my study was to investigate the complex relationship between badgers, their
habitat, and roads within south-central BC. To my knowledge, few studies have attempted to
summarize and interpret the patterns of mesocarnivores in relation to roads (Clarke et al.
1998, Grilo et al. 2009), with none relating to the North American badger. My intent was to
use this information to provide natural resource and transportation planners and managers
with insight and tools that may be used to mitigate road mortality for badgers in south-central
BC, and ideally, elsewhere in the species’ range.
The specific objectives of this thesis are presented in three main chapters. In the
current chapter, I provide background and the historical context for my study, including a
detailed description of the study area (see below). In Chapter 2, I report on the ecology of
these animals in this region, mainly from the perspective of roads and sources of mortality
and then examine the implications of this for the badger population. Finally, in Chapter 3, I
present more focused discussion on how an understanding of the ecology of badgers
(particularly movements) may be integrated into habitat protection and restoration plans,
including mitigation for current and future road infrastructure projects, thereby reducing
impacts on this species.
STUDY SITE DESCRIPTION
My field work was conducted in the Cariboo region surrounding the community of
100 Mile House in south-central BC (N51° 39΄ 11˝, W121° 17΄ 29˝). Non-invasive data
collection coupled with an intensive radio-telemetry study occurred over an area
encompassing approximately 7,200 km2 (Figure 1.3). Due to the vast expanse and
remoteness of this area the radio-telemetry component occurred within the core area (local
study area) where highways bisected the regional study area.
The Cariboo region in BC is predominantly a broad, flat plateau between two
mountain systems consisting of level to gently rolling landscapes, with incised river valleys
and uplands locally rising above the general surface (Steen and Coupé 1997). The study area
consists primarily of the physiographic subdivisions of the eastern Fraser Plateau, with a
small amount of the Fraser Basin in the north, and very small areas of the Quesnel Highlands

11

Figure 1.3. Regional badger study area (outer) and local study area (inner) with
highway locations near the community of 100 Mile House (N51° 39΄ 11˝, W121° 17΄
29˝) in south-central British Columbia, Canada.

12

to the east and Thompson Plateau to the south (Holland 1976). The Fraser Plateau, bounded
by the Coast Mountains to the west and Columbia Mountains to the east, covers the majority
of the study area with elevations of 900 – 1,500 m.
The area is underlain primarily by flat-lying to gently dipping olivine basalt bedrock,
which in most areas is covered by a mantle of medium- to coarse-textured glacial till. In
some areas the till is shaped by glaciers into drumlins and eskers (Holland 1976, Steen and
Coupé 1997). A small portion of the study area is made up of the Fraser Basin where it
extends north along valley bottoms from Lac La Hache to Williams Lake. This area is
characterised by a low-relief, low-lying plain covered by glacial till and glacial lake deposits.
The till frequently forms eskers and drumlins (Figures 1.4A and 1.4B).
The climate of the region is influenced by three principal air masses: warm, moist
Pacific air form the west; cold, dry Arctic air from the north; and warm, dry Great Basin air
from the south (Steen and Coupé 1997). Since the region is in the lee of the Coast Mountains,
the moist westerly airflow is restricted resulting in a much drier continental climate. As
Pacific air moves eastward across the Fraser Plateau, humidity and precipitation increase.
Mean annual precipitation at 100 Mile House is 453.3 mm and at nearby Bridge Lake,
approximately 44 km southeast, it is 599.3 mm (Environment Canada climate normals,
unpub. data). The Arctic air mass lies north of the region during most of the summer but
influences climate during the winter months, resulting in periods of very cold temperatures.
However, the region is somewhat buffered by the full effects of the cold by mountain ranges
to the northeast. High snowfall events often occur when the cold Arctic air invades the region
and interacts with the moister Pacific air. Northern higher elevation parts of the region are
more affected by Arctic air incursions than are southern portions and, as a result, the northern
areas experience colder snowier winters and cooler summers. For example, mean annual
temperature at Lillooet (elev. 198 m, approx. 115 km southwest) is 9.2° C, while at 100 Mile
(elevation 1059 m) it is 4.2° C (Environment Canada climate normals, unpub. data). Warm
Great Basin air has relatively little effect on the climate in this region, except in the Fraser
River valley south of the Chilcotin River confluence, where hot, dry air in the summer can
penetrate from the south, resulting in high daytime temperatures and clear skies. The
relatively high elevations of the Fraser Plateau usually restrict the influence of this hotter air
mass in the study area, resulting in cooler summer temperatures.

13

Figures 1.4A and 1.4B. Aerial photo of Highway 97 bisecting valley bottom habitat in
northern portion of study area (above); and rolling hills with grassland/pasture habitat
in central portion of study area (below) in south-central BC, Canada.

14

Vegetation patterns in the study area reflect the patterns of climate, topography and
soils. The majority of the Fraser Plateau landscape is made up of climax stands of relatively
open and slow growing lodgepole pine (Pinus contorta) . Higher precipitation areas and sites
adjacent to wetlands often include hybrid white spruce (Picea glauca var. engelmanii).
Where elevations become lower and summer temperatures are higher, Douglas-fir
(Pseudotsuga menziesii) forests are well represented on the landscape often in mixtures with
lodgepole pine. In these drier sites, Douglas-fir is the climax tree species. Pinegrass
(Calamagrostis rubescens) is consistently present and often dominates the undergrowth of
these forests. Further eastward, as precipitation increases, the Douglas-fir forests become
seral to spruce forests, which become increasingly common. Small pockets of deciduous
forests, dominated by aspen (Populus tremuloides), occur throughout most of the Fraser
Plateau being more common and extensive in the northern Fraser Basin area towards
Williams Lake. In the study area, native grasslands occur extensively on the slopes of the
Fraser River valley south of the confluence of the Chilcotin River. These grasslands extend
east into the Fraser Plateau before gradually transitioning to conifer dominated forests, with
small grasslands occurring locally on dry, south-facing slopes. Grasslands are also common
in open forest parklands in parts of the Interior Douglas-fir (IDF) zone further eastward.
Wetlands are very common in the study area outside of the Fraser River valley, and are
predominantly fens, marshes, and shrub-carr [(Steen and Coupé 1997) (Figures 1.5A and
1.5B)].
Soil parent materials in the Cariboo are predominantly glacial till of medium to
coarse texture (Valentine and Dawson 1978). The till materials were generally transported
only relatively short distances by Pleistocene glaciers and, as a result, the composition of the
till often reflects the local bedrock. Coarse textured tills are common across the western
Fraser Plateau where they were derived from granitic rocks of the Coast Mountains.
Glaciofluvial and fluvial deposits occur locally, forming outwash plains, terraces, and eskers
especially in valley bottoms. Lacustrine deposits occur locally and sporadically, particularly
in the north portion of study area near Williams Lake. Deep organic deposits have a small
extent and are much localized.

15

Figures 1.5A and 1.5B. Examples of badger habitat adjacent to wetlands. Field
technician documenting a burrow in fine-textured soils upslope of a wetland, with
grassland transitioning to Douglas fir dominated forests in background (top); Typical
badger habitat adjacent to wetlands in the central portion of study area (below).

16

LITERATURE CITED
Apps, C. D., N. J. Newhouse, and T. A. Kinley. 2002. Habitat associations of American
badgers in southeastern British Columbia. Canadian Journal of Zoology-Revue
Canadienne De Zoologie 80:1228-1239.
Banks, P. B. 1999. Predation by introduced foxes on native bush rats in Australia: do foxes
take the doomed surplus? Journal of Applied Ecology 36:1063-1071.
Blanco, J. C., Y. Cortes, and E. Virgos. 2005. Wolf response to two kinds of barriers in an
agricultural habitat in Spain. Canadian Journal of Zoology-Revue Canadienne De
Zoologie 83:312-323.
Boitani, L. 2001. Carnivore introductions and invasions. Pages 123-144 in J. Gittleman, S.
M. Funk, D. W. Macdonald, and R. K. Wayne, editors. Carnivore Conservation.
Cambridge University Press, Cambridge, United Kingdom.
Buskirk, S. W., and W. J. Zielinski. 2003. Small and mid-sized carnivores. Pages 207-249 in
C. J. Zabel, and R. G. Anthony, editors. Mammal community dynamics :
management and conservation in the coniferous forests of western North America.
Cambridge University Press, New York.
Cardillo, M., A. Purvis, W. Sechrest, J. L. Gittleman, J. Bielby, and G. M. Mace. 2004.
Human population density and extinction risk in the world's carnivores. Plos Biology
2:909-914.
Carroll, C., R. F. Noss, and P. C. Paquet. 2001. Carnivores as focal species for conservation
planning in the Rocky Mountain region. Ecological Applications 11:961-980.
Ceballos, G., P. R. Ehrlich, J. Soberon, I. Salazar, and J. P. Fay. 2005. Global mammal
conservation: What we must manage? Science 309:603-607.
Clarke, G. P., P. C. L. White, and S. Harris. 1998. Effects of roads on badger Meles meles
populations in south-west England. Biological Conservation 86:117-124.
Crooks, K. R., and M. E. Soule. 1999. Mesopredator release and avifaunal extinctions in a
fragmented system. Nature 400:536-566.
Donadio, E., and S. W. Buskirk. 2006. Diet, morphology, and interspecific killing in
Carnivora. The American Naturalist 167:524-536.
Duquette, J. F. 2008. Population ecology of badgers (Taxidea taxus) in Ohio. M.S. Thesis,
The Ohio State University, Columbus.
Eldridge, D. J. 2004. Mounds of the American badger (Taxidea taxus): Significant features of
north American shrub-steppe ecosystems. Journal of Mammalogy 85:1060-1067.

17

Eldridge, D. J., and W. G. Whitford. 2009. Badger (Taxidea taxus) disturbances increase soil
heterogeneity in a degraded shrub-steppe ecosystem. Journal of Arid Environments
73:66-73.
Ferreras, P., J. J. Aldama, J. F. Beltran, and M. Delibes. 1992. Rates and causes of mortality
in a fragmented population of Iberian lynx Felis-pardina Temminck, 1824. Biological
Conservation 61:197-202.
Forman, R. T. T., J. Bissonette, A. Clevenger, C. Cutshall, V. Dale, L. Fahrig, C. Goldman,
K. Heanue, J. Jones, D. Sperling, F. Swanson, T. Turrentine, and T. Winter. 2003.
Road ecology : science and solutions. Island Press, Washington, DC.
Fraser, D. F. 2000. Species at the edge: the case for listing of “peripheral” species. Pages 49–
53 in Proceedings of Conference on the biology and management of species and
habitats at risk, 1999. British Columbia Ministry of Environment, Lands and Parks,
Victoria, and University College of the Cariboo, Kamloops, British Columbia,
Canada. Volume 1:49–53.
Gittleman, J. L., and M. E. Gompper. 2005. Plight of the predators: The importance of
carnivores for understanding patterns of biodiversity and extinction risk. Pages 370388 in P. Barbosa, and I. Castellanos, editors. Ecology of Predator-Prey Interactions.
Oxford University Press, Oxford (United Kingdom).
Grilo, C., J. A. Bissonette, and M. Santos-Reis. 2009. Spatial-temporal patterns in
Mediterranean carnivore road casualties: Consequences for mitigation. Biological
Conservation 142:301-313.
Hairston, N. G., F. E. Smith, and L. B. Slobodkin. 1960. Community structure, population
control, and competition. American Naturalist 94:421-425.
Henke, S. E., and F. C. Bryant. 1999. Effects of coyote removal on the faunal community in
western Texas The Journal of Wildlife Management 63:1066-1081.
Holland, S. S. 1976. Landforms of British Columbia: a physiographic outline. 2nd ed. B.C.
Dep. Mines, Mineral Resources, Victoria, B.C. Bull. No. 48.
Hoodicoff, C., and R. Packham. 2007. Cariboo Region Badger Project: Year End Report
2006-07. BC Ministry of Environment, 100 Mile House, BC. Report Technical, 38 pp
+ appendices.
Hoodicoff, C., S. 2003. Ecology of the badger (Taxidea taxus jeffersonii) in the Thompson
Region of British Columbia: Implications for Conservation. Thesis (M.Sc. Biology),
University of Victoria, August 2003.
Hoodicoff, C. S., K. W. Larsen, and R. D. Weir. 2009. Home range size and attributes for
badgers (Taxidea taxus jeffersonii) in south-central British Columbia, Canada.
American Midland Naturalist 162:305-317.

18

jeffersonii-Badger-Recovery-Team. 2008. Recovery Strategy for the Badger (Taxidea taxus)
in British Columbia. Prepared for the B.C. Ministry of Environment, Victoria, B.C.
45 pp.
Jones, K. L., D. H. Van Vuren, and K. R. Crooks. 2008. Sudden increase in a rare endemic
carnivore: Ecology of the island spotted skunk. Journal of Mammalogy 89:75-86.
Kinley, T. A., and N. J. Newhouse. 2008. Ecology and translocation-aided recovery of an
endangered badger population. Journal of Wildlife Management 72:113-122.
Kirkpatrick, M., and N. H. Barton. 1997. Evolution of a species' range. American Naturalist
150:1-23.
Kyle, C. J., R. D. Weir, N. J. Newhouse, H. Davis, and C. Strobeck. 2004. Genetic structure
of sensitive and endangered northwestern badger populations (Taxidea taxus taxus
and T-t. jeffersonii). Journal of Mammalogy 85:633-639.
Lampe, R. 1982. Food habits of badgers in east central Minnesota. Journal of Wildlife
Management 46:790-795.
Lesica, P., and F. W. Allendorf. 1995. When are peripheral populations valuable for
conservation. Conservation Biology 9:753-760.
Lindzey, F. G. 1978. Movement patterns of badgers in northwestern Utah. Journal of
Wildlife Management 42:418-422.
Long, C. A., and C. A. Killingley. 1983. The badgers of the world. C.C. Thomas, Springfield,
Ill., U.S.A.
Maron, J. L., J. A. Estes, D. A. Croll, E. M. Danner, S. C. Elmendorf, and S. L. Buckelew.
2006. An introduced predator alters Aleutian Island plant communities by thwarting
nutrient subsidies. Ecological Monographs 76:3-24.
Meserve, P. L., D. A. Kelt, W. B. Milstead, and J. R. Gutiérrez. 2003. Thirteen years of
shifting top-down and bottom-up control. Bioscience 53:633-646.
Messick, J. P. 1987. North American badger. Pages 586-597 in M. Novak, J. A. Baker, M. E.
Obbard, and B. Malloch, editors. Wild furbearer management and conservation in
North America. Ontario Trappers Association, and Ontario Ministry of Natural
Resources,Toronto, Canada.
Messick, J. P., and M. G. Hornocker. 1981. Ecology of the badger in southwestern Idaho.
Wildlife Monographs 76:1-53.
Michener, G. R. 2004. Hunting techniques and tool use by North American badgers preying
on Richardson's ground squirrels. Journal of Mammalogy 85:1019-1027.

19

Minta, S. C. 1993. Sexual differences in spatiotemporal interaction among badgers.
Oecologia 96:402-409.
Newhouse, N. J., and T. A. Kinley. 1999. Update COSEWIC status report on the American
badger (Taxidea taxus) in Canada. Committee on the Status of Endangered Wildlife
in Canada.
_____. 2001. Ecology of badgers near a range limit in British Columbia. Technical report to
Columbia Basin Fish & Wildlife Compensation Program, Nelson, B.C.; and Parks
Canada Agency, Radium Hot Springs, B.C.
Noss, R. F., H. B. Quigley, M. G. Hornocker, T. Merrill, and P. C. Paquet. 1996.
Conservation biology and carnivore conservation in the Rocky Mountains.
Conservation Biology 10:949-963.
Paine, R. T. 1980. Food Webs - linkage, interaction strength and community infrastructure the 3rd Tansley Lecture. Journal of Animal Ecology 49:667-685.
Paulson, N. J. 2007. Spatial and habitat ecology of North American badgers (Taxidea taxus)
in the native shrubsteppe ecosystem of Eastern Washington. Thesis (M.Sc. in natural
resource science), Washington State University, August 2007.
Polis, G. A., A. L. W. Sears, G. R. Huxel, D. R. Strong, and J. Maron. 2000. When is a
trophic cascade a trophic cascade. Trends in Ecology and Evolution 15:473-475.
Proctor, M. F., B. N. McLellan, C. Strobeck, and R. M. R. Barclay. 2005. Genetic analysis
reveals demographic fragmentation of grizzly bears yielding vulnerably small
populations. Proceedings of the Royal Society B-Biological Sciences 272:2409-2416.
Quinn, J. H. 2008. The ecology of the American badger (Taxidea taxus) in California:
assessing conservation status on multiple scales. Ph. D., University of California
Davis.
Rahme, A. H., A. S. Harestad, and F. L. Bunnell. 1995. Status of the badger in British
Columbia.
Ray, J. C. 2005. Large carnivores and the conservation of biodiversity. Island Press,
Washington.
Ripple, W. J., and R. L. Beschta. 2004a. Wolves and the ecology of fear. Bioscience 54:755766.
Roemer, G. W., M. E. Gompper, and B. Van Valkenburgh. 2009. The ecological role of the
mammalian mesocarnivore. Bioscience 59:165-173.
Sargeant, A. B., and D. W. Warner. 1972. Movements and denning habits of a badger.
Journal of Mammalogy 53:207-&.

20

Sovada, M. A., J. M. Roaldson, and A. B. Sargeant. 1999. Foods of American badgers in
west-central Minnesota and southeastern North Dakota during the duck nesting
season. American Midland Naturalist 142:410-414.
Steen, O. A., and R. A. Coupé. 1997. A field guide to forest site identification and
interpretation for the Cariboo Forest Region. BC Ministry of Forests. Victoria, BC.
Taylor, I. 2005. A carnivore's inquiry. New York Times Book Review:32-32.
Valentine, K. G., and A. B. Dawson. 1978. The Interior Plateau. Pages 121-134 in K. G.
Valentine, P. N. Sprout, T. E. Baker, and L. M. Lavkulich, editors. The Soil
Landscapes of British Columbia. Agriculture Canada & B.C. Ministry of
Environment.
van Woudenberg, A. M., and D. A. Christie. 1997. Flammulated owl (Otus flammeolus)
population and habitat inventory at its northern range limit in the southern interior of
British Columbia. Pages 466-476 in Biology and Conservation of Owls of the
Northern Hemisphere.
Vucetich, J. A., and T. A. Waite. 2003. Spatial patterns of demography and genetic processes
across the species' range: Null hypotheses for landscape conservation genetics.
Conservation Genetics 4:639-645.
Warner, R. E., and B. VerSteeg. 1995. Illinois badger studies. Division of Wildlife
Resources, Illinois Department of Natural Resources, Illinois, USA.
Weir, R. D., H. Davis, and C. Hoodicoff. 2003. Conservation strategies for North American
badgers in the Thompson and Okanagan regions. Habitat Conservation Trust Fund,
Artemis Wildlife Consultants, Armstrong, B.C. Report Final, pp 103.

21

CHAPTER 2 – ROAD ECOLOGY OF BADGERS (TAXIDEA TAXUS) IN SOUTHCENTRAL BRITISH COLUMBIA, CANADA
INTRODUCTION
Over the past decades the density of transportation routes in North America has
increased significantly, resulting in large numbers of wildlife being killed on roads (Stoner
1925, Lalo 1987). Understanding the factors that contribute to road mortality has emerged as
one of the more critical issues facing wildlife conservation in many developed countries
(Davies et al. 1987, Jones 2000, Mumme et al. 2000, Dique et al. 2003). Roads and their
traffic have major impacts on animal populations and communities, and traffic mortality is a
serious conservation concern worldwide for many species, especially those with declining or
restricted distributions such as rare carnivores (Bennett 1991, Trombulak and Frissell 2000,
Forman et al. 2003). The field of road ecology has emerged as a powerful tool for
understanding the complex relationship between wildlife and the effects of roads on their
populations.
Many carnivores have been seriously impacted by the expansion of transportation
systems and networks; however we know little about carnivore response to the extent and
magnitude of traffic mortality (Smith and C. K. Dodd 2003, Grilo et al. 2009). Carnivore
populations exhibit several traits that can render them particularly vulnerable to habitat
fragmentation and highway impacts. Because of their relatively large home-range sizes,
carnivores often have to cross one or more highways to fulfill their food or water
requirements, find mates, or disperse into unoccupied habitats (Ruediger 1998). Carnivore
populations can be put at further risk when mortality rates rise because of their low
population densities and reproductive rates (Ruediger 1998, Riley et al. 2006). Roads have
been identified as a serious source of mortality for many medium-sized carnivores (or
‘mesocarnivores’) as their smaller size makes them more difficult to detect by motorists, and
mortality rates of these animals from collisions often is underestimated (Ferreras et al. 1992,
Clarke et al. 1998, Riley et al. 2006)
As mesocarnivores, North American badgers (Taxidea taxus) appear highly
susceptible to road mortality. In particular, near the northwestern mountainous portions of
their range a combination of factors predisposes the animal to road mortality: (1) highways

22

in this region are constructed in valley bottom habitats preferred by badgers, (2) large home
ranges are correlated with extensive movements, with peak movement periods (mating
season) coinciding with peak summer traffic volumes; (3) roadside right-of-ways are
attractive to badger prey, and (4) badgers are most active at night, when drivers have greater
difficulty seeing a smaller animal on the highway. In BC, the leading cause of mortality for
badgers has been identified as road mortality on major highways, with rates of ranging from
13% to an astounding 86% (Weir et al. 2003, Kinley and Newhouse 2008, Hoodicoff et al.
2009). This phenomenon is not restricted or new to BC, with American wildlife biologists in
the northwest reporting elevated badger mortality due to collisions with vehicles (Messick
1987), and similar impacts are seen on populations of the European badger (Meles meles)
(Clarke et al. 1998). Further habitat loss and degradation coupled with highway and traffic
expansion will only serve to increase road mortality of these animals.
In south-central BC, the effects of road mortality on the North American badger
become even more serious when viewed within the context of the animals being at the
northern periphery of their range (see Figure 1.1 in Chapter 1). Indeed, the extreme northern
limit for the badger subspecies T.t. jeffersonii is found within the central portion of BC,
Canada (N51° 39΄ 11˝, W121° 17΄ 29˝). Peripheral populations such as this often represent
strongholds for species undergoing declines, and may be important reservoirs of genetic
diversity (Lesica and Allendorf 1995). In many cases, broad landscape approaches may be
efficient in protecting peripheral populations (Noss 1987, Franklin 1993), but variations in
life-history and demographics often make detailed information on these peripheral
populations critical, especially for maintaining genetic variability (Millar and Marshall 1992,
Soulé and Mills 1998, King and Burke 1999). Anthropogenic impacts on populations likely
will work in an additive or synergistic manner with ecological constraints already occurring
at the limit of a species’ range.
As mentioned, the mountainous topography in central BC renders badgers prone to
road mortality, partly because of the distribution of habitat. In this region, the open dry
grasslands typically associated with these animals (Lindzey 1978, Messick and Hornocker
1981, Goodrich and Buskirk 1998, Paulson 2007) are predominantly found in valley bottoms
where transportation corridors are typically placed. However, data collected during previous
home range studies in BC have revealed a wider range of habitats used by badgers than

23

expected. For example, the animals have been shown to use higher-elevation forests,
including recent forest harvest blocks where prey (e.g., ground squirrels) were abundant
(Apps et al. 2002, Weir et al. 2003). At the extreme northern range limit of the jeffersonii
badger, there is a strong contrast in both topography and climate, suggesting different
patterns of habitat use may be in effect. At this location, the topography is generally less
rugged with patchy burrowing soils, and the climate is considerably harsher due in part to
both increased elevation and latitude. Initial burrow searches and badger sighting records in
this region indicate that badgers inhabit a diverse array of habitats, ranging from bunchgrass
to sub-boreal pine and spruce zones. However, road mortality also occurs in the region,
suggesting that the formulation of effective management plans for these animals will likely
need to encompass a wider range of habitat types than previously thought, as well as address
the incidence of road mortality. Given that traffic volume likely will increase in the face of
northern development, a solid understanding of the ecology of these animals, including the
impact and causes of road mortality, is needed to ensure their persistence at the northern
periphery of their range.
I conducted a detailed ecological study of the northern-most badger population in BC.
The overarching goal was to better understand the relationship of these animals to their
extreme northern environment, and how that was potentially linked to road mortality on
major highways bisecting the area. This complex relationship was investigated using
conventional radio-telemetry in tandem with newly emerging GPS technology. The specific
goals of my study were to:
examine home range and movements of badgers, including dispersal of juveniles,
particularly in the summer months when road mortality is highest;
document reproductive success of adult females including seasonal timing of kit
emergence and natal den abandonment and how roads may affect these patterns;
quantify mortality factors of the studied population;
document the distribution and abundance of animals in the region and the influence of
roads on these parameters; and
determine how existing highway infrastructure features may influence badger
movements and connectivity.
Herein I use my data to examine the demographic parameters and the ecological
effects of roads pertinent to the development of management plans for this extreme northern

24

population, as well as badger populations elsewhere that may be facing similar challenges.
Given the strong tie of my study to road ecology, my field work was focused towards the
summer season (highest traffic volumes, breeding, etc.). A simultaneous study investigated
the winter denning ecology of the animals (Symes 2013).
STUDY AREA
From 2007 to 2010, I conducted field work in the area surrounding the community of
100 Mile House, BC (N51° 39΄ 11˝, W121° 17΄ 29˝) (Figure 2.1). Non-invasive data
collection coupled with an intensive radio-telemetry study occurred over an area
encompassing approximately 7,200 km2. Due to the vast expanse and remoteness of this area
the radio-telemetry component occurred within the core area (local study area) where
highways bisected the regional study area. The study area is comprised of a broad, flat
plateau between two mountain systems, covered by glacial till and glacial lake deposits with
elevations ranging from 900 – 1,500 m above sea level. The topography is level to gently
rolling landscapes with incised river valleys and uplands locally rising above the general
surface (Steen and Coupé 1997). The till frequently forms eskers and drumlins that provide
badgers with friable soils suitable for burrowing.
Moisture regimes increase from west to east across the plateau. Summers are
influenced by warm, moist Pacific air from the west and warm, dry Great Basin air from the
south, resulting in a drier continental climate. Winters are influenced by Arctic air masses,
leading to very cold winter temperatures with relatively low winter precipitation.
Much of the study area is dominated by Douglas-fir (Pseudotsuga menziesii) and
lodgepole pine (Pinus contorta) trees that occur in open savannah-like stands on drier sites,
with bluebunch wheatgrass (Pseudoroegneria spicata) and rough fescues (Festuca spp.)
forming the understory. Wetter sites at higher elevations support dense conifer stands with
pinegrass (Calamagrostis rubescens), and feathermoss (Hylocomium splendens) understory.
Soopolallie (Shepherida canadensis), kinnikinnik (Arctostaphylos uva-ursi), rose
(Rosa spp.), and twinflower (Linnaea borealis) are common understory shrubs. At lower
elevations, the west and south of the study area is bordered by very dry warm bunchgrassopen forest communities, while at higher elevations to the north and east it is replaced by the

25

Figure 2.1. Regional study area (outer) and local study area (inner) boundaries with highway locations near the community
of 100 Mile House (N51° 39΄ 11˝, W121° 17΄ 29˝) in south-central British Columbia, Canada.

26

cool, dry climates of lodgepole pine and spruce (Picea glauca) climax forests (Steen and
Coupé 1997). Trembling aspen (Populus tremuloides) stands are common, and aspen is often
scattered throughout the forest stands in the study area. Small pocket grasslands occur locally
on south-facing slopes often bordering the many lakes and wetlands in the region. For a more
detailed description of the study area refer to Chapter 1.
There are two major highways that pass through the study area: Highway 97 bisects
the study area in a north-south direction, and Highway 24 runs on the eastern portion of the
study area in an east-west direction (see Figure 2.1). Much of Highway 97 follows the
historical wagon route along the valley bottom through native grasslands and open forests
between the communities Cache Creek and Quesnel. With the growing need for an improved
transportation system from the west coast to the economically-expanding north, Highway 97
will be upgraded to accommodate this growth. These upgrades are projected to include the
widening of a 440 km portion through the central interior to four lanes of traffic. By 2022,
almost 50 per cent of this segment will be either three or four lanes wide [(British Columbia
Ministry of Transportation and Infrastructure [MoTI] 2009)]. Highway 24 currently serves
primarily as a two-lane tourist highway linking many fishing lakes to smaller communities in
the interior of the province.
Traffic volumes along Highway 97 range from 3,737 vehicles/day annually (AADT)
to a peak of almost 7,000 vehicles/day during summer, while Highway 24 has much lower
traffic volumes with an approximate average annual daily traffic volume of 911 vehicles/day,
and a high of 2,000 vehicles/day in summer [(British Columbia Ministry of Transportation
and Infrastructure [MoTI] 2010)].
METHODS
Live-capture and radio-telemetry
Active ground searches by vehicle and on foot were conducted within the local study
area to determine recent badger activity (i.e. fresh burrows). Attempts were made to capture
an equal number of badgers both within close proximity to highways (≤ 1 km) and distant
from highways (≥ 4 km). Badgers were live-captured during May to August of 2007 – 2009
using Victor # 1 ½ padded soft-catch foot-hold traps (Victor-Oneida, Woodstream Inc.,
Lititz, Pa.), anchored with 40 – 45 centimeter cable T-anchor and set at active burrow

27

entrances. Traps were set overnight and checked twice daily, at least every 12 hours.
However, if traps were set at a burrow known to be occupied, the site was monitored
continuously until capture or closing of the traps.
Captured badgers first had their weight visually estimated for drug volume and then
were noosed using a restraint pole. Telazol

(tiletamine and zolazepam) was hand-injected

in a large muscle mass at a dose of 5 – 10 mg/kg body weight, in a concentration of 100
mg/mL. Once immobilization had taken place, body condition was assessed, vital signs were
monitored, and ophthalmic ointment was applied.
Immobilized badgers were either transported to a veterinary clinic or a mobile
veterinary field unit for intraperitoneal implantation of VHF radio-transmitters using standard
sterile surgical techniques. The animals were transported in a ventilated 200 L plastic barrel
with a lid and secured in the back of an open-aired vehicle. An internal temperature sensor
was used to remotely monitor temperature inside the barrel. Biological samples and
morphometrics were conducted post-surgery while the animal was still under anaesthesia.
Hair and tissue samples were collected for genetic analysis. No teeth were extracted for aging
as we classified age by tooth wear (i.e., juvenile, young adult, mature adult, old adult). In
order to distinguish individuals from public sightings (photo evidence), field observations,
camera traps, and road mortality, we photographed the unique facial pelage and patterns of
all individuals (including kits that we did not radio-tag).
Badgers were held for 1 – 2 hours after surgery until they were deemed sufficiently
recovered from anaesthesia, demonstrating normal balance, posture, reflexes, and reactions.
Once recovered, the animals were transported to the initial point of capture and released at
their burrow. Monitoring then occurred regularly over the next 48 hours to ensure that the
badgers returned to regular activity patterns.
Radio-telemetry locations were obtained by using a vehicle-mounted omni-directional
antenna and an R-1000 receiver (Communication Specialist Inc., Orange, Calif.) to locate the
general area the animal was using, and then following the signal on foot using a handheld
radio-telemetry system. If precise locations of the animal (e.g. in a burrow, hunting in a
pasture) could not be determined, I took several compass bearings and Universal Transverse
Mercator (UTM) coordinates from a global positioning system (Garmin, Inc.) for

28

triangulation. The triangulation software program LOCATE ІІІ (Pacer Computing,
Tatamagouche, NS) was used with the Maximum Likelihood Estimator (MLE) to determine
an accurate location from triangulation data. If the animal was stationary, I located the
burrow it was in and recorded UTM coordinates, broad habitat type, presence/absence of
Columbian ground squirrels (Spermophilus columbianus) within a detectable distance by
surveyors (≤ 100 m), and geophysical characteristics. Locations were obtained from early
morning throughout the day until dusk. Late night locations in the dark were rarely attempted
as the animals were usually active and moving and I did not want to deter or influence their
foraging behaviour or movement patterns. During summer field season (May to August) my
crew and I attempted to collect a minimum of one location per week per individual, but this
frequency shifted to once every two weeks in the fall and winter months when badgers
reduced their movements, sometimes entering into a period of inactivity or torpor (see Symes
2013). Attempts were made while radio-tracking badgers to observe and document any road
crossings, either above or below grade. Periodic fixed-wing telemetry flights were conducted
when an individual was not located for two or more consecutive weeks. Locations of
individuals identified during previous DNA sampling sessions also were used to augment
radio-telemetry locations of study animals.
A subset of study animals were outfitted with GPS datalogger backpacks to determine
detailed movement patterns. To accomplish this, telemetered animals were recaptured > 1
year after implantation using the same capture and handling techniques described above.
Animals were fitted with a nylon harness that encircled their neck and chest and was
stabilized by two crossing sternum straps that prevented the top-heavy GPS unit from sliding
under the animal (Figures 2.2A and 2.2B). The GPS and harness were secured with a quickrelease device so that the entire unit could easily be retrieved from the animal after being
deployed for 2 – 4 weeks. The attachment harness was pilot-tested on a tame, captive badger
prior to deployment on wild animals. The GPS unit paired with a small VHF tracking device
(Sirtrack Ltd., NZ) was attached to the top of the harness in order to obtain exposure to
satellites in the sky when the animal was moving above ground (Figure 2.3). The GPS unit
was programmed to obtain a fix every 15 minutes. If satellites were detected by the unit
(animal was above ground), the satellite search would continue for three minutes until a fix
was obtained; otherwise, the unit would shut off until the next programmed fix.

29

Figures 2.2A and 2.2B. Harness system testing on tame, captive badger illustrating
sternum strap for stabilization of GPS datalogger backpack (top), and removing the
harness system (bottom) at British Columbia Wildlife Park, Kamloops BC, Canada,
June 2008.

30

GPS datalogger
VHF unit

Figure 2.3. Dorsal view of a wild free-ranging male badger outfitted with GPS datalogger nine days after deployment exiting
from a burrow near Meadow Lake, BC, 14 August 2009. Photo taken by remote motion camera.

31

If no satellites were detected (e.g., animal below ground), the unit would continue the
satellite search for 30 seconds before shutting down until the next programmed fix. This
programming prevented undue drainage on the battery caused by prolonged satellite searches
by the unit while the animal was underground.
Home range and movements
To calculate badger home ranges I applied fixed kernel (FK) estimation using leastsquares cross validation (Seaman and Powell 1996; Powell 2000; Kernohan et al. 2001).
Kernel estimators have a number of features that make them useful for home ranges
calculations: they work well with small amounts of data, they are robust to autocorrelation,
they are nonparametric, and they allow multiple centers of activity rather than a simple home
range outline (Kernohan et al. 2001). I did not use the adaptive kernel method as it tends to
perform poorly, often over-estimating home range areas (Seaman and Powell 1996, Powell
2000). I used the program The Home Ranger (Hovey 1999) to calculate FK home range
estimates followed by post-processing and importing the outputs into Arcview 3.2a (ESRI)
for further spatial analysis. Annual home ranges (95% FK) were calculated for individuals
with > 30 radio-locations spanning a year of radio monitoring. Due to tracking constraints
and the unpredictable road mortality of study animals, seasonal range sizes were sometimes
estimated using ≥ 5 radio-locations.
A geographic information system (GIS) was used to display 95% FK, including 100%
minimum convex polygons (MCP) using the extension Home Range in ArcView (Rodgers
and Carr 1998) for all animals. Displaying 100% MCP in concert with 95% FK is useful for
displaying locations with disjunct kernels (concentrated areas of use) derived by the fixed
kernel (FK) method with an overall area that an animal may use (MCP).
I divided seasonal home ranges based on four intervals that corresponded to annual
life-history phases of badgers. Radio-telemetry locations were then grouped into seasonal
categories based on these periods of times: summer for rearing of kits by females and
breeding for both sexes (1 June – 31 August), fall for post-breeding recovery and pre-winter
fattening for both sexes (1 September – 30 November), winter for reduced activity and
dormancy for both sexes (1 December – 28 February), and spring for parturition/rearing for
females (1 March – 31 May). Where sufficient samples sizes were collected, comparisons of

32

adult male and female annual and seasonal home range sizes were conducted using MannWhitney (U) Rank Sum Tests. Using FK methods, home range core areas (50% utilisation
distribution, UD) were delineated to compare distances from the edge of core areas to the
nearest highway.
Movement data acquired from GPS datalogger units were analysed for animals that
collected data for > 24 hr period in order to be considered independent after harnesses were
initially outfitted. Movement rates were divided into 2 categories: (1) an active period that
included movements collected for every programmed 15 minute fix up to 2 hours to account
for short term underground forays, presumably foraging for fossorial rodents, and (2) an
inactive period that spanned periods where badgers made minimal movements for extended
periods of time (i.e., > 2 hrs).
The systematic collection of hair from badgers to identify individuals using DNA
fingerprinting began in 2003 to 2006 by government biologists (Hoodicoff and Packham
2006), and concluded with my data collection for this thesis from 2007 to 2009. This system
provided both distribution and abundance data in the region. A similar approach has been
used successfully with other mustelid species including American marten (Martes
americana), fishers (Pekania pennanti), wolverines (Gulo gulo), and European badgers
(Meles meles) (Belant 2003, Zielinski et al. 2006, Mulders et al. 2007, Scheppers et al. 2007).
Data collected using this method provided me with additional information for tracking
individual movements, identifying road mortality individuals, and kit dispersal patterns.
Whenever possible, a combination of DNA samples, telemetry, dispersal locations
and known fates of individual badgers were determined to provide insight into movement
patterns. Several hundred DNA samples (n = 618) collected across the regional study area
from 2004-2006 prior to my radio-telemetry study provided baseline demographic data.
During this period, 51 badgers (23 ♀, 28 ♂) including two litters, were identified in the
region (Hoodicoff and Packham 2007). This DNA hair-snagging program was continued
during my live-trapping program in 2007 – 2009. As juvenile badgers have been known to
remain on their mother’s territory for one year before dispersal, dispersal distances were
defined by the distance from the known natal den to the furthest point of detection (e.g.,
radio-telemetry, DNA, road mortality) from the den within 24 months. Documenting these

33

dispersals allowed me to compare dispersal distances between male and female juveniles (ttests).
Reproduction
Monitoring of telemetered adult females was conducted more frequently in March to
determine the time and location that natal burrows were excavated and/or used. If a female
remained at a single burrow for more than two consecutive weeks during late March, a
remote motion camera was placed near the den entrance to document the activity pattern of
the female, number of kits, and time of kit emergence from the natal den (Figure 2.4). This
method also documented the time when natal dens were abandoned and new maternal dens
were established. Kits were individually recognized by their unique facial pelage patterns,
allowing me to distinguish individual kits and thereby calculate litter size (at least at time of
emergence). Facial pattern recognition also allowed me to identify other adult badgers that
visited the burrow complex during camera deployment. Observations of copulation by
telemetered badgers (both females and males) allowed me to confirm the timing of breeding.
Mortality
Telemetry allowed me to determine mortality factors and location, particularly during
the summer (high traffic) season. I also solicited sightings of badgers killed on roads from the
public through local newspaper releases, radio interviews, and by placing information posters
on community billboards (Appendix C). Reports of dead badgers were verified as soon as
possible and biological samples, UTM coordinates, occupancy of ground squirrels in vicinity,
and associated roadside habitat types were collected at confirmed mortality sites. Previous
badger road mortality sites and archived DNA samples collected from 2003 – 2006 by
government biologists also were used in the analysis.
Known badger road mortalities were spatially defined using fixed kernel (FK)
methods that highlighted areas of concentrated road mortality sites or ‘hotspots’. These road
mortality hotspots were delineated using the 70% utilisation distribution (UD) kernel to
identify segments of highway where badger road mortality most commonly occurred.
Roadside habitats were then determined at road mortality sites in the study area from
Predictive Ecosystem Mapping (PEM) conducted for the Cariboo region (2008).

34

Figure 2.4. Remote infrared motion camera housed in security box attached to stand deployed at badger burrow.

35

Kaplan-Meier survival curves (LogRank test) were used to assess the survival
probability for radio-tagged males and females over the duration of the study.
Finally, home range core areas (defined by 50% UD) of tagged badgers were
delineated and distances from highways to the edge of core areas were compared to assess
distance from roads and survival rates in relation to road mortality events for both sexes.
Distribution and abundance
Tracking individuals using DNA
We deployed hair-snagging devices throughout the regional study area where badgers
were known to occur, at burrows that had been documented either through direct
observations, radio-telemetry, and reports submitted from the public, or from previous
burrow surveys conducted by government biologists. In order to maintain some form of
systematic consistency, hair-snags were deployed during the same seasons and in areas that
were previously sampled from year to year across the regional study area.
Over the course of the study, different material was used to snag hair, ranging from
barbed wire to Velcro® pads to pinned knaplock (i.e., carpet edging material) attached to a
metal band. In all cases, two opposing snagging plates (approximately 2.5 cm × 4 cm) were
riveted to curved metal strapping that had snagging material affixed to it. These devices were
anchored to the roof of a burrow using two metal ardox (spiral) nails that prevented slippage
out of the soil roof, or by attaching a small gauge wire to the device and then feeding it up
through a hole to the surface above the burrow, where it was fastened to a wooden peg
anchor (Figures 2.5 & 2.6). This non-invasive method did not deter badgers from entering or
exiting the burrow, allowing for the assumption of equal catchability to be met.
Live-capture samples, snagged hair, and road mortality tissue samples were air dried,
stored in paper envelopes, and sent to Wildlife Genetics International (Nelson, BC) for DNA
extraction and analysis. Over the course of collecting and analysing DNA for this study there
were no changes to the DNA extraction protocol or to the convention used of recording data
during extraction (QIAGEN DNeasy tissue extraction kit: QIAGEN Inc., Valencia,
California). Samples were pre-screened with a single marker to identify species so that only

36

Figure 2.5. Schematic diagram (l) of hair-snagging device deployed in badger burrow
and close up photo of hair-snagger (r) with suitable hair sample for DNA analysis. Each
of the two snagging pinned knaplock segments were approximately 2.5 cm x 4 cm in
dimension.

Figure 2.6. Hair-snagger deployed in active badger burrow revealing suitable hair
sample for DNA analysis. Photo taken near 100 Mile House, BC by P. Verkerk.

37

badger samples were brought forward for further genetic analysis (Paetkau pers. comm.
2010). A standard set of 7 microsatellite markers was used to generate a complete 7-locus
genotype with minimal clean-up procedures. Error-checking was done using a protocol that
virtually eliminates the possibility of recognizing false individuals through genotyping error
(Kendall et al. 2009). We used the markers ZFX/ZFY/SRY that produced a high-confidence
gender analysis (Paetkau pers. comm. 2010).
Population modeling
Survival, capture, and recovery
Three modeling methods were used to determine (1) demographic parameters, (2)
population size and (3) population growth. The program MARK (version 6.1) was unable to
calculate weighted averages for some demographic parameters (i.e., survival, recapture)
using one method alone. In addition to the Burnham (Both) model to estimate demographic
parameters, models POPAN and Pradel‐λ generated estimates of population size (N), and
growth (λ) through program MARK (White and Burnham 1999, Cooch and White 2000).
I used a multi-model selection procedure to estimate demographic parameters for the
regional badger population, based on the recovery of marked individuals. More specifically,
the information-theoretic approach described by Burnham and Anderson (1998) was used to
evaluate the relative support of multiple models describing relationships between population
parameter variables, such as survival, capture probabilities, and recovery probabilities.
Population parameters can be determined by using various complementary sources of data
(e.g., recoveries, recaptures, and telemetry), thereby allowing estimates of parameters to be
calculated when they would not otherwise be available. This approach improves the
precision of parameter estimates, and provides “flexibility” in accommodating encounter data
that are collected “opportunistically” throughout the year (Cooch and White 2000). For
example, the assumption of permanent or random emigration in a study based on live
recaptures can be relaxed if live re-sightings or dead recoveries are included in the analysis.
There were four main steps in the estimation of demographic parameters for badgers
in the regional study area. Firstly (1), I constructed encounter histories for marked
individuals, then (2) I used the Burnham (Both) model to estimate survival (S), recapture (p),
and recovery (r) probabilities, followed by (3) determining how well the resulting general

38

model fit the data by comparing the deviance of the general model using the bootstrap
method and correcting for poor fit, which then (4) allowed me to use a maximum-likelihood
method to determine the reduced model that best represented the data.
Burnham (1993) developed an approach for combining dead recovery and live
encounter data into a single analysis, and I followed this approach using program MARK. I
first constructed encounter histories for each individual badger from both live-recapture data,
and live and dead encounters (Cooch and White 2000). Individual badgers were initially
identified through live-capture and/or DNA sampling protocols. Following this method, a
badger was recorded as “recaptured” in the annual summer field season if it was detected
through any of the various methods being used, or as “not captured” if it was undetected
during that time period. Once a badger was detected, subsequent recaptures in that same
season were not considered in the analysis. DNA ‘captures’ were removed if they originated
outside of the season or beyond the geographic extent of the study area. Badgers were
analyzed separately according to gender.
Using the encounter histories constructed for individual badgers, I determined
estimates of survival (S), recapture (p), recovery (r), and fidelity (F) using the Burnham
(Both) model for live and dead encounters for open populations (Burnham 1993). For the
joint analysis of live encounter and dead recovery data, MARK uses the Seber (‘S and r’)
parameterization primarily to take advantage of increased flexibility (Cooch and White
2000). Based on the assumption that recoveries of dead marked animals and recaptures will
occur in the same sampling area, fidelity (F) for the marked sample is fixed at ‘1’ (Cooch and
White 2000).
A general (global) model with all the population parameters mentioned above,
dependent on group (i.e., male or female) and time, was fit to the Burnham (Both) model,
followed by deriving models with reduced parameters to determine the best representation of
the data. The fit of the general model was determined by comparing its deviance to the mean
deviance of 1000 bootstrap-replicated models. I found this general model deviated
significantly from the mean bootstrap deviance (P = 0.045), indicating poor fit. Accordingly,
model selection was then based on a quasi-likelihood corrected version of Akaike’s
Information Criterion (QAICc)(Cooch and White 2000).

39

In order to determine the model that best represented the data, model selection was
conducted using maximum-likelihood methods (Burnham and Anderson 2001). The
corrected Akaike Information Criterion (AICc) was used to determine the model that was the
most parsimonious (Burnham and Anderson 2004). To account for any overdispersion in the
data (goodness of fit test), the model deviance of the AICc (ĉ=1) was adjusted to the
deviance of the general model (ĉ=1.174), as calculated by the median ĉ test in MARK. The
adjusted ĉ required that model selection be conducted using the quasi-likelihood estimator,
QAICc (Burnham and Anderson 1998). Population parameter probabilities were then
estimated using weighted averages from the QAICc model weights. In order to reduce errors
deriving from invalid estimates of some parameters and standard errors during the modelaveraging process, all models with less than one percent support from the QAICc were
dropped from consideration (Cooch and White 2000). Weighted parameter estimates then
were obtained from the reduced model list.
Population size
Given that data collection spanned several years, permanent additions or deletions to
the population were present, so I used the open population model POPAN to estimate
population size (N). This is a Jolly-Seber model derivative that uses mark-recapture rates to
derive estimates of abundance and related parameters (Schwarz and Arnason 1996). This
process varies from the Burnham (Both) model in that it estimates population size in addition
to survival (S) and capture (p) probabilities. However, this model was not used to determine
S and p simply because Program MARK is unable to calculate weighted averages for the
POPAN models, where it will do so for the Burnham (Both) Jolly-Seber models. Final

selection was based on the three most parsimonious models from the Burnham (Both) JollySeber model (i.e., models with over 10% support from the QAICc). Constructed models were
compared using the AICc to determine the most parsimonious model, which was then used to
estimate population size for badgers in the region from 2003-2008.
Population growth
Population growth (λ) was estimated using the Pradel‐λ formulation of the
Jolly‐Seber model, which also provides an estimate of survival (S), similar to the parameters
survival (Φ), and recapture (p) (Pradel 1996). Similar to the POPAN formulation model, the

40

Pradel‐λ model does not allow calculation of weighted averages for Φ and p, and therefore it
could not provide estimated values for them. Models were therefore constructed using the
three most parsimonious models from the Burnham (Both) Jolly‐Seber analysis. Thus, λ was
allowed to vary with time or it was held constant to determine which reduced model best fit
the data. Model selection was achieved using AICc to determine the most parsimonious
model from which the population growth estimate could be obtained. The Pradel-λ estimator
requires that at least three sampling periods be carried out in order to calculate variables
(Pollock et al. 1990). Unfortunately for this analysis, DNA identification cannot determine
age class and all of the badgers had to be treated as a single group for the calculation of λ.
Highway structures influencing movements and mortality
Previous projects reported that existing underpass structures (i.e., culverts) may be
used by badgers to cross under major transportation corridors (Newhouse and Kinley 2001). I
therefore collected data on 114 potential underpass structures that existed within the local
study area where badgers were known to occur, as determined by radio-tagged animals, road
mortality sites, and burrows in the vicinity. This was done to determine the degree to which
suitable and unsuitable crossing structures were present at badger road mortality sites. I
recorded these structures as being suitable if a badger could easily traverse the structure
unimpeded and unsuitable for a crossing if they showed blockages (e.g., crushed culvert
ends, infilled or blocked openings, etc.), or if they had been originally installed with a
hanging end (i.e. culverts protruding from the highway bank ≥ 45 cm above ground), or if
water or mud ≥ 5 cm in depth was present in the actual structure.
Of the 114 structures that I investigated, 101 were corrugated metal culverts ranging
in size from 400 mm to 1300 mm diameter, 12 were corrugated metal and concrete box
livestock underpasses (2000 mm – 4000 mm diameter), and there was a single corrugated
metal pipe installed for a stream crossing (4800 mm). Two open spans (i.e., bridges) were
encountered during culvert surveys but were not assessed as they contained running water
across their entire width.
To document the frequency and ratio of above and below grade road crossings by
badgers, remote motion cameras were placed at entrances to underpass structures. This
permitted a comparison of GPS movement data with camera data. Over the course of three

41

field seasons, I monitored 41 potential underpass structures using remote motion cameras
(Reconyx Inc., RC55) within radio-tagged badger home ranges. In order to simultaneously
sample as many concurrent underpass structures as possible within a badger’s home range, a
single camera was placed at one opening of an existing structure and each motion event
triggered three successive photos with no time delay between events. Successful crossings by
animals were then determined by examining the sequential images from the cameras to
determine if crossings were successful or not. Underpass structures ranging in size from 400
mm to 4000 mm were in the sample (37 corrugated metal culverts + 4 concrete box
underpasses). Culverts ranging in size from 500 – 700 mm were most common along
highway segments accounting for 80% of the sample surveyed. I selected three highway
segments for this intensive monitoring where both radio-tagged badgers and road mortality
occurred: (1) 3.2 km of Highway 24 in eastern portion, (2) 17.6 km of Highway 97 in central
portion, and (3) 7.2 km of Highway 97 in northern portion of study area. These sites were
spread across the study area ranging from two- to four-lanes in width.
To determine if the frequency of existing suitable crossing structures affected badger
road mortality rates, I compared the number of suitable and non-suitable crossing structures
between two segments of Highway 97 where known badger road mortalities frequently
occurred. These segments had comparable traffic volumes, habitat types, soils, and landscape
characteristics. In addition, small mammal surveys were conducted during August 2009
along highway right-of-way at these sites to determine the presence and distribution of
common prey species. The north segment was 8.9 km and south segment was 10.1 km in
length and were separated by 35 km. Using the Fisher’s exact test, I compared the two
segments for suitable versus unsuitable underpass crossing structures (i.e., culverts) and
differences in mortality among those highway segments.
RESULTS
Despite the ongoing loss of study animals to road mortality, I was able to collect
sufficient data to portray the ecology of badgers in this region, and the factors contributing to
road mortality. A total of 84 badgers (42 ♂, 38 ♀, 4 unknown) were detected in the region
during the course of this study. Of this sample, 16 animals (5 ♂, 11 ♀) were captured in the
local study area, radio-tagged, and tracked from 2007 – 2009. A subset of 5 study animals (3

42

♂, 2 ♀) were equipped with a GPS datalogger backpack system for short-term intensive
monitoring, this being to my knowledge the first such use of this technology on the species.
Live-capture and radio-telemetry
From 2007 to 2009 I captured more than twice as many females than males (5 ♂, 11
♀); this sample may reflect the sex ratio of the population given the propensity for males to
suffer mortality on roads (see below). Attempts to capture an equal number of badgers in
close proximity to (≤ 1 km), and distant (≥ 4 km) from highways, resulted in seven and nine
captures respectively. I collected 1,057 independent radio-locations spanning 914 days (July
31, 2007 – January 31, 2010). Individuals were located at least once a week from May –
August (xˉ = 4.2 days, SD = 0.81) and two to three times a month during spring, fall, and
winter (xˉ = 8.52 days. SD = 1.32).
Home range
The average annual home range size for both sexes was 70.1 km2 (SD = 91.2, n = 11,
Table 2.1). Across all seasons, adult male home ranges were considerably larger those for
adult females (overall, t = -27.8, n = 9, 2, P = 0.001). Although five males were tagged, three
were struck and killed on the highway within three months of tagging, providing limited
spring and summer data on habitat use. Due to these constraints, the data from these animals
were not included in the annual home range analysis, although seasonal home ranges were
calculated using the radio-telemetry data collected during the summer and spring field
seasons. The average home range for the two males that provided sufficient numbers of
locations (415 and 716 days, see individuals BM01 and BM02, Table 2.1) was 253.1 km2
(SD = 9.8, n = 2) while average for females were 29.4 km2 (SD = 12.2, n = 9).
Important differences were noted in the size and shape of home ranges for male
versus female badgers, and how they were orientated in relation to highways. In general, all
badgers had some type(s) of road within their home ranges, with 12 home ranges containing
a major highway, three having at least one paved secondary road, and one having a few twolane gravel roads. Home ranges for males were generally linear in nature following the
corridors of suitable habitat (Figure 2.7). Male animals north of the town of 100 Mile House
generally inhabited the valley bottom, resulting in Highway 97 bisecting the majority of their

43

Table 2.1. Annual and seasonal* 95% fixed kernel home ranges (95% FK) for all study animals, July 30, 2007 - January 31,
2010. Sample sizes for each seasonal estimate are provided in the 5 rightmost columns. Seasons are defined as follows: Spring
(March-May), Summer (June-August), Fall (September- November), and Winter (December-February).
Fixed Kernel 95% (km2)
Gender

♀

ANNUAL

SPRING

SUMMER

FALL

27.3
32.0
32.5
19.6
22.2
12.6
46.7
49.4
22.6
29.4 (12.2)

18.2
23.5
12.0
11.0
10.6
12.4
8.6
13.8 (5.2)

15.2
19.4
18.8
13.7
17.7
9.0
53.2
15.5
47.8
21.3
23.2 (14.9)

4.4
9.5
7.6
8.1
5.0
7.6
6.2
7.9
5.3
21.8
13.8
8.8 (5.0)

MEAN ♂ (SD)

246.1
260.0
253.1 (9.8)

114.2
65.0
22.9
26.7
57.2 (42.5)

270.2
235.0
103.1
61.8
147.1
163.4 (87.7)

POP. MEAN (SD)

70.1 (91.2)

29.5(32.2)

69.9 (83.8)

Identity
BF01
BF02
BF03
BF04
BF05
BF06†
BF07
BF08
BF09†
BF10
BF11
MEAN ♀ (SD)

♂

Number of radio-locations

BM01
BM02
BM03
BM04
BM05

† Individual assumed to be less than one year old

WINTER

AN

SP

SU

FA

WI

1.6
2.0
1.6
2.1
1.1
2.7
5.9
2.2 (1.6)

119
91
118
68
118
42
50
43
35
33
32

26
11
22
14
12
13
12
2
1
2
0

41
47
48
24
52
4
12
30
24
20
21

29
24
25
20
30
14
15
8
8
9
6

23
9
23
10
24
11
11
3
2
2
5

72.2
42.4
57.3 (21.1)

35.2
18.0
26.6 (12.2)

147
85
31
23
22

33
11
6
7
0

59
58
25
16
22

32
11
0
0
0

23
5
0
0
0

16.3 (19.7)

7.8 (11.6)

44

Figure 2.7. Adult male badger fixed kernel (FK 95%) polygons and 100 minimum
convex polygons (100% MCP, straight lines) home ranges (n=5) in south-central BC,
Canada, 2007 – 2010.

45

linear home ranges. Changes in topography south of 100 Mile House resulted in a mosaic of
pothole lakes and wetlands, where badger habitat was linked to glacio-lacustrine features,
such as drumlins and eskers. These areas were characterised by a mosaic of wetlands, pocket
grasslands and aspen copses, timber harvested sites, and dry open forests of lodgepole pine.
These habitats were linked in a more west-east orientation where Highway 97 intersects this
smaller area, resulting in less road mortality risk.
Female home ranges were generally less linear in nature, and due to their smaller size
were spread across the region in areas of suitable high-quality habitats (i.e., high quality
native grasslands, Figure 2.8). Many of these habitats were in close proximity to both
Highway 97 and 24 surrounding the 100 Mile House area. Only four of 11 female home
ranges did not include highways bisecting their core home ranges (50% UD), with the edge
of the core area an average distance of 12.0 km (SD = 4.4, n = 4)) from highways as
compared to other females (n = 7) that had core areas < 500 m from highways (xˉ = 71 m, SD
= 18, P = 0.012).
Movements
Over the course of this study I collected 503 radio-locations during summer, 231
during fall, 151 during winter, and 172 during spring for an average of 6.3 days/location (SD
= 1.7). Badgers moved much more in summer than all other seasons (Table 2.2). Average
seasonal movement rates estimated from radio monitoring for all badgers were 840 m/day
(SD = 847) in summer, 287 m/day (SD = 155) in fall, 105 m/day (SD = 100) in winter, and
356 m/day (SD = 483) in spring. Males had higher movement rates than females for all
seasons but were significantly higher during summer and spring (P < 0.05). Reproductive
females (n = 6) had significantly lower movement rates in spring than females that did not
have litters (n = 5) that season (P < 0.05).
Monitoring of tagged badgers showed that 15 of 16 animals had home ranges in close
proximity to major paved roads, with 12 of 16 animals known to frequently cross highways,
with all five males crossing highways at least twice throughout the summer season. A
combination of radio-telemetry and DNA burrow records showed one male traveled a linear
distance of 132 km over 14 days (23 June – 09 July, 2008), moving from a ranch in the north
to a pasture in the central portion of the study area and back again. Another telemetered male

46

Figure 2.8. Female badger home ranges (n = 11) 95% FK polygons and 100% MCP
(straight lines) in south-central BC, Canada, 2007 – 2010.

47

moved a minimum of 152 linear kilometers along the Highway 97 corridor over 17 days (04
– 21 August 2008) averaging 30.5 km between successive radio-locations. This same male
showed consecutive daily movements ranging from 1.9 km to 10.4 km (n = 7), with a longest
movement recorded of 64.1 km over a duration of 3 days (14 – 17 August, 2008).
Of the 11 adult females radio-tagged, only four were not recorded crossing at least
one of the highways. These four females had 50% utilisation distribution core areas ranging
from ≥ 6 – 16 linear kilometers away from highways.
GPS backpacks were deployed on five study animals (3♂, 2♀) during the 2008 and
2009 summer seasons, with three animals providing reliable detailed movement data (> 24
hrs – Table 2.3). One GPS datalogger unit was lost when the bonding agent attaching the
GPS to the harness failed (animal BM01 – not shown on Table 2.3), while another animal
(BM03, Table 2.3) managed to remove his harness within 20 hours of deployment.
For those badgers that provided >24 hrs of GPS data collection (n = 3), the average
rate of movement recorded was 4,800 m/day (SD = 1.1). During the summer period (July and
August) these badgers moved more extensively between GPS fixes during the dusk to dawn
period (21:00 – 06:00, xˉ = 185 m, P < 0.05) than the diurnal period (06:00 – 21:00, xˉ = 94
m).
The average rate of movement calculated from GPS fixes for badgers that were active
was 188 m (SD = 19.3) every 19 min and 48 sec (SD = 02 min, 11 sec), providing an
estimate of 564 m/hr when actively moving above ground. When badgers were inactive or
GPS fix attempts failed, presumably because animals were underground, the average
movement rate dropped to 61.3 m (SD = 54.5) every 03 hours and 15 min (SD = 02 min, 06
sec) resulting in approximately 19 m/hr.
Based on the GPS data, badgers were capable of moving on average 6,800 m over 12
hours when actively traveling and foraging, and < 228 m over 12 hours when inactive. The
longest period that a badger was inactive was 48 hours, as demonstrated by an older male
(BM02). The longest periods of inactivity for females (n = 2) were 20 hr: 50 min (BF05) and
12 hr: 16 min (BF01). The longest distance traveled by a badger between GPS fixes (15
minutes) was 1,014 meters, a travel speed of > 4 km/hr. Within the GPS sample, there were
no significant differences (t = -0.59, P = 0.61) in movements detected between genders.

48

Table 2.2. Average movement rates (m/day) derived from conventional radio-tagged
badger data monitored between 2007 and 2010 for all seasons in south-central BC.
Season
Category

Summer

Fall

Winter

Spring

All Badgers (♀,♂)

840 (847)

287 (155)

105 (100)

356 (483)

Male ♂

1640 (1163)*

421 (100)

197 (223)

930 (563)*

Female ♀

476 (261)*

263 (153)

85 (60)

78 (67)*

Reproductive ♀

373 (77)

254 (152)

96 (79)

33 (20)*

Non-reproductive ♀

598 (358)

274 (172)

70 (29)

135 (58)*

* P < 0.05

Table 2.3. Summary of GPS datalogging deployment on four badgers in south-central
BC, July - August 2008 and 2009.

# of
fixes

Total
distance
traveled
(m)

Distance
traveled/day
(m)

Animal
ID/ sex

Date deployed

Year

Time collecting
data
(hrs:min:sec)

BM03 ♂

22 - 24 Jul

2008

19:35:40

39

5,886

-

BF01 ♀

20 Jul - 07 Aug 2008

73:35:54

104

11,495

3,749.3

BF05 ♀

04 - 17 Aug

2009

217:07:23

277

54,513

6,025.7

BM02 ♂

06 - 22 Aug

2009

298:30:24

399

57,515

4,626.5

49

The detailed GPS data collected from one female (BF05) with kits indicated that, in at
least some cases (e.g. summer), radio-telemetry was significantly underestimating the
amount of road-crossings that an individual badger conducted over a given time period.
Conventional radio-telemetry of this adult female was conducted while she was outfitted with
a GPS datalogger, 04 - 17 August. During this time, five radio-telemetry fixes were collected
on the female, all of which were on the south side of the highway suggesting she did not
cross the highway during that time period. However, the GPS data revealed that during this
time she moved extensively, averaging 251 m/hr (SD = 323), and crossing the highway six
times. All of these crossings occurred from late evening until early morning (Figure 2.10).
Remote motion camera monitoring of culverts in the vicinity of her crossings did not detect
badger underpass use during this period, suggesting all crossings were above the road grade.
Dispersal
DNA samples from individual badgers live-captured, hair-snagged at burrows, or
carcasses collected from roads confirmed several matches from previous hair-snagging
sessions prior to my research. Dispersal movements detected several young animals (n = 9)
crossing highways to occupy new territories, with a minimum estimate of 50% road mortality
for dispersing juveniles (Table 2.4). Average detected minimum dispersal distance was 43.2
km (SD = 32.1) for both sexes. Dispersal distances averaged 53.5 km (SD = 36.2, n = 6) for
males, while females averaged 30.8 km (SD = 24.3, n = 5; t = 1.18, P = 0.133). One juvenile
male (BM04) was documented making an extensive movement over the course of a year,
travelling 63 kilometers southwest across Highway 97 during dispersal, then moving 86
kilometers north in less than two months the following summer, once again crossing the
highway before being radio-tagged in the northern portion of the study area. The straight line
distance from the natal den to his last known location was 83 kilometers.
Reproduction
The 13 natal burrows that I detected provided data on reproductive success of the
females, in terms of minimum litter size and successful kit emergence (Appendix A). The
average number of kits I detected emerging from the den was 2.4 (SD = 0.76, n = 13). For
those natal dens that had a complete history of camera monitoring, the mean date of kit
emergence for all years was 16 May (SD = 9.5, n = 8); after the first detected emergence by

50

Figure 2.9. Detailed GPS track of adult female badger's (BF05) movements crossing Highway 24 near Bridge Lake, BC, 04 17 August 2009. Squares represent sequential GPS locations acquired every 15 minutes. Inset map: Entire GPS track
overview and home range outline (95% FK) derived from VHF radio-telemetry.

51

Table 2.4. Minimum detected dispersal distances of badger kits as determined by DNA
or live-capture mark/recapture at burrows. Detection method describes whether the
animal was live-captured or DNA hair-snagged at burrow.
Mother

Study

ID

Kit ID

Year

(DNA)

(DNA)

2005

F179

2006

2008

2008

All

F507

F507

-

Known

animal

Detection

dispersal

Known fate (as of

Sex

ID

method

(km)

Jan 31, 2010)

-

♀

-

Natal den

-

Unknown

F316

♀

BF04

Burrow snag

67

Road mortality

M326

♂

-

Burrow snag

21

Unknown

M355

♂

-

Burrow snag

68

Road mortality

F328

♀

BF11

Live capture

1.2

Alive

-

♀

BF03

Natal den

-

Alive

F534

♀

BF02

Live capture

20

Road mortality

M520

♂

-

Burrow snag

20

Road mortality

M513

♂

BM04

Live capture

90

Road mortality

F533

♀

-

Burrow snag

38

Unknown

-

♀

BF03

Natal den

-

Alive

F898

♀

BF08

Live capture

28

Alive

M884

♂

BM06*

Live capture

24

Alive

-

♀

-

Maternal den

-

Unknown

M931

♂

-

Burrow snag

-

Unknown

M932

♂

-

Burrow snag

98

Road mortality

♂,♀

* Study animal tagged post-2010.

= 43.2

52

the kits, an average of 22 days (SD = 3.8. n = 8) expired before natal den abandonment
occurred.
Interestingly, in spring 2010, one adult female occupying a home range where ground
squirrels were absent had a litter of two kits emerge approximately 2 – 3 weeks behind that
observed for all other females. Based on occurrence surveys of Columbian ground squirrels
during telemetry tracking, 12 of 13 natal dens had ground squirrels ≤ 100 m and the
reproductive output (i.e., number of kits emerging from den) of mature females in areas
containing colonies of fossorial rodents were notably higher (xˉ = 2.67 kits, n = 9) than that
for the mature female (BF10) in the area lacking colonial rodents (xˉ = 1.0, n = 1). Younger
breeding females reproductive output was higher in areas containing ground squirrels than
the area lacking this consistent prey base, but lower than mature experienced females (xˉ =
1.25, n = 3).
Mortality
No incidents of ‘natural’ mortality (i.e., predation) or intentional killing by humans
were detected through the telemetered cohort (n = 16); rather, all known mortalities to study
animals arose from vehicle collisions on paved roads (n = 8). Eight of 16 (4 ♂, 4 ♀) tagged
animals were struck and killed on Highway 97, which accounted for 80% of males and 36%
of females in my sample. All of the males died in July and August, whereas females were
killed over a broader time span (Appendix B). Kaplan-Meier survival curves reveal a
significant difference in survival time for radio-tagged badgers between genders (χ2 = 4.528,
P = 0.03) with an average survival time of 728 days for females and 329 days for males
(Figure 2.11).
From 2004 to 2010, reports and retrievals of 39 badgers (16 ♂, 18 ♀, 5 unknown)
killed on roads were investigated and mapped, with 72% of these deaths occurring on the
major highway in the region (Hwy 97), and only 3% occurring on the secondary highway
(Hwy 24). The remaining 25% occurred on secondary paved 2-lane roads (≥ 80 km/hr speed
limit) throughout the region.
Over the course of the study, females whose core home ranges (50% UD) spanned
highways had a 43% survival rate (n = 7) whereas females isolated from highways had 100%
survival rate (n = 4; t stat = -5.36, df = 9, P = <0.001). The core areas for all radio-tagged

53

1.0

Female
Male

Survival

0.8

0.6

0.4

0.2

0.0
0

200

400

600

800

1000

1200

Time (days)

Figure 2.10. Kaplan-Meier survival curves for male and female radio-tagged badgers
that died on roadways in south-central BC, 2007-2009.

54

male badgers were an average distance of 1.82 km (SD = 2.6, n = 5) from highways, with
these animals having only a 20% survival rate.
Females and their dependent kits residing near major roadways were also subject to
road mortality. After one female (BF07) moved her kits in early summer to a maternal den
adjacent to the highway, she was struck and killed, with one kit subsequently being hit 13
days later (the fate of the second kit was unknown). Road mortality was detected for two
more juveniles from two separate litters, representing a total of 50% kit mortality from these
three litters. One of these juveniles was observed being struck and killed by a vehicle while
following their mother across a paved road. Additionally, juveniles are exposed to road
mortality during dispersal as family units identified from DNA hair-snagging (see Table 2.4)
had 50% road mortality of dispersing juveniles (4 ♂, 2 ♀) from four litters (n = 12).
From 2004 – 2010, badger deaths on Highway 97 most often occurred from May to
December with a peak in July (Figure 2.12). Males accounted for 40% of road deaths in July
and August, which coincided with a peak in summer traffic volume and the breeding season,
while female deaths were spread throughout their active seasons.
Three separate road segments along Highway 97 (approx. 18 km) in the northern
portion of the study area (100 Mile House to Williams Lake) contributed to the majority of
badger-vehicle collisions, which accounted for 11% of the paved highway through the
regional study area (Clinton to Williams Lake, 165 km). The 70% utilisation distribution of
badger road mortality highlights three areas that contributed to over half (60%) of all the road
mortalities in the region. These include a 4 km radius from 100 Mile House (includes
secondary paved roads), an 8 km segment of Highway 97 in the central portion of the study
area, and a 6 km segment of Highway 97 in the northern portion of the study area (Figure
2.13). Areas that consisted of non-forest upland with improved pastures were the dominant
roadside habitat-type associated with badger road mortalities (n = 19).
Of the eight telemetered animals lost on roads from 2008 - 2010, only three were
reported by the public (38%) with the remaining five mortalities confirmed by radiotelemetry monitoring. Conservative estimates of the percentage of the regional population
killed by the highway annually range from 7 – 11% based only on the reporting rate of road
mortalities and increase to 15 – 25% when adjusted for the corrected reporting rate.

55

Figure 2.11. Percentage of badger-vehicle collisions in relation to average daily traffic volume (ADT)* by month on
Highway 97 in south-central British Columbia, Canada, 2004 – 2010. *Source: Ministry of Transportation and Infrastructure,
Traffic Data Program, 2004 – 2008.

56

Figure 2.12. Plotted road mortality locations with 70% fixed kernel utilization
distribution contours revealing three ‘roadkill hotspots’ in relation to major roads in the
northern portion of the study area in south-central BC, 2004 – 2010. Sixty percent of
badger road mortality in the region occurred in these three zones along Highway 97 in
the study area from 2004 – 2010.

57

Population modeling
DNA hair-snagging
Over the course of the study, I collected 1,004 samples of hair from different
burrows. The average success rate of hair samples having suitable DNA for extraction and
thus being individually identified was 49% (SD = 18.8) over six annual sampling sessions
(range: 22% - 75%). This approach was therefore successful at first-time identification of
individuals as well as providing ‘re-capture’ data.
Population parameters (Survival, recapture, and recovery)
Thirty-nine badgers were recovered dead from roads during the course of the study.
Twenty of the dead badgers were included in the live and dead encounter histories for the
DNA sampling component of the study, while the remaining 19 were excluded as they were
collected post-2008, beyond the years used for analysis.
The general population model generated through MARK was S(t*g)p(t*g)r(t*g)F(t*g)
where S = the survival probability as a function of the interaction between year (t) and sex
(g), p = capture probability as a function of the interaction between year and sex, r =
recovery probability as a function of the interaction between year and sex, and F = fidelity
probability as a function of the interaction between year and sex. This model was used to test
goodness of fit to the data by summarizing the discrepancy between observed and expected
values (P = 0.045). Based on the measure of fit of the model to the data, the variance
inflation factor was estimated as ĉ = 1.174 (SE=0.161). A suite of candidate models were
then generated and ranked according to QAICc (Table 2.5). The most parsimonious model
was S (year) p(constant) r(constant) F(constant), where S = survival probability as a function of year
with the remaining variables held constant
Sex did not appear to be a major factor in influencing survival from year to year, but
the survival probability varied by year for both genders combined (Table 2.6). Survival
probabilities ranged from a low of 0.45 (SE = 0.28) in 2008 to 0.78 (SE = 0.10) in 2005. The
probability of ‘capturing’ badgers over the course of the study was 0.75 (SE = 0.06) and the
probability of recovering marked dead animals was 0.28 (SE = 0.08).

58

Table 2.5. Suite of top-ranked candidate models derived by the Reduced Burnham
(Both) Jolly-Seber method. These models provides estimates for survival (S), recapture
(p), and recovery (r) of marked individuals based on corrected Quasi-Akaike
Information Criterion (QAICc) with an adjusted ĉ=1.174. Model names indicate if
survival (S) or probability of capture (p), recovery (r), and fidelity (F) were held
constant (.), or varied with group (g) or time (t). ∆QAICc is the difference in QAICc
values between a model and the most parsimonious model. ‘Model weights’ indicates
the support for a model. Number of parameters specifies the number of variables
estimated for a model. QDeviance is the deviance of a model from the full general
model.
Model

QAICc

∆ QAICc

Model

Model

Number of

Weights

Likelihood Parameters

QDeviance

S(t)p(.)r(.)F(.)

256.59

0.00

0.212

1.000

7.00

100

S(.)p(.)r(.)F(.)

256.76

0.17

0.195

0.917

3.00

109

S(t)p(g)r(.)F(.)

256.91

0.33

0.180

0.850

8.00

98

S(g)p(.)r(.)F(.)

258.75

2.16

0.072

0.339

4.00

108

S(g)p(g)r(.)F(.)

259.09

2.50

0.061

0.286

5.00

107

S(t)p(g)r(g)F(g)

259.18

2.59

0.058

0.274

9.00

98

S(t)p(g)r(g)F(.)

259.18

2.59

0.058

0.274

9.00

98

S(.)p(.)r(.)F(t)

259.23

2.64

0.057

0.267

6.00

105

S(.)p(g)r(.)F(.)

259.63

3.04

0.046

0.218

5.00

107

S(.)p(.)r(.)F(g)

261.02

4.44

0.023

0.109

5.00

109

S(t*g)p(.)r(.)F(.)

261.32

4.74

0.020

0.094

12.00

93

S(t*g)p(g)r(.)F(.)

261.55

4.97

0.018

0.083

13.00

91

59

Table 2.6. Output from program MARK vers. 6.1 for most parsimonious model
providing population parameter estimates for survival by year (S), probability of
recapture (p), and probability of recovery of dead marked individuals (r) for the
regional badger population in south-central BC, Canada, 2003 – 2008.

BADGER DEAD-LIVE RECOVERY
Standard Error and Confidence Intervals Corrected for ĉ = 1.174
Ranked 1st
Real Function Parameters of {S(t)p(.)r(.)F(.)}
95% Confidence Interval
Parameter

Year/gender

Estimate SE

Lower

Upper

---------------------

------------------------

----------- -----

-------- --------

1: (S) survival

2003/both sexes

1.00

0.00

1.00

1.00

2: (S) survival

2004/both sexes

0.67

0.13

0.39

0.86

3: (S) survival

2005/both sexes

0.77

0.10

0.52

0.92

4: (S) survival

2006/both sexes

0.62

0.10

0.41

0.79

5: (S) survival

2007/both sexes

0.71

0.11

0.46

0.87

6: (S) survival

2008/both sexes

0.45

0.28

0.08

0.88

7: (p) recapture

All years/both sex

0.75

0.06

0.61

0.85

8: (r) recovery

All years/both sex

0.27

0.08

0.15

0.46

9: (F) fidelity

Fixed (see methods)

1.00

0.00

1.00

1.00

60

Population size
The POPAN model provided derived estimation of abundance. Based on this, the
population size in the regional study area was estimated to have increased over the course of
the study (Figure 2.14). The minimum number of badgers detected alive in 2003 was 12 with
the overall population estimate increasing to 72 (SE = 3.5, 95% CI = 67-83) by 2008. This
provides an estimated density of one badger/100 km2 across the regional study area.
Population growth
Using the Pradel-λ method, a suite of candidate models were calculated and ranked in
order of lowest scores for AICc (Table 2.7). The most parsimonious model for the Pradel-λ
population growth model was λ = Φ (constant) p(constant) λ (constant). The resulting population
growth rate (λ) was 1.249 (SE = 0.07). Both the POPAN population estimate and Pradel- λ
growth model generated in MARK provide similar trajectories in population growth,
suggesting an increasing population of badgers in the regional study area (Table 2.8).
Highway structures influencing movements and mortality
Motion cameras deployed on 41 underpass structures allowed me to document 108
occurrences of badgers crossing underneath highways. Size of structures used by badgers
ranged from 500 mm to 2500 mm diameter. The most frequently-used underpass structure
was 500 mm corrugated pipe culverts that accounted for 81% of all badger crossings, with
livestock underpasses (> 2 m) accounting for only 8% of these crossings (Figure 2.15).
Although 500 mm culverts had the more frequent use, they were also the most common
culvert size in the sample. Deployment of cameras on underpass structures ranged from 35 –
520 days (xˉ = 115 days, SD = 84.6) resulting in the detection of badgers crossing under
roads at 37% (SD = 0.48, n = 15) of the structures monitored.
The average number of badger crossings per structure monitored from 2007 – 2009
was 3.58 (SD = 6.74) with a range of 0 – 31 crossings/structure. The average number of
camera nights/ badger crossing was 67 (SD = 2, n = 4,525) with July having the highest
frequency of crossings. There were 274 badger detections recorded with 108 documented
crossings (39%). Non-crossings were detected 166 times (61%) where badgers were recorded
near the opening of an underpass but not detected passing through them. These situations
included badgers entering the underpass then promptly exiting < 1 minute, investigating

61

100

Est. Badger Population (N)

80

60

40

20

0
2003

2004

2005

2006

2007

2008

Figure 2.13. POPAN model derived population estimate of Year
badgers across the regional study area in south-central British
Columbia, Canada, 2003-2008. The 2003 estimate is based on minimum number of badgers detected alive. Bars represent +/- 1
Col 1 vs Col 2
standard error.

62

Table 2.7. Pradel-λ badger population model selection based on Akaike Information
Criterion (AIC). Model names indicate if population growth (lamda, λ) was held
constant (.) or varied by time (t). Model weight indicates the support for a model.
Number of parameters specifies the number of variables estimated for a model.
Pradel‐λ model does not allow calculation of weighted averages for Φ and p, and
therefore it does not provide estimated values for them.
Model

AICc

Delta

AICc

Model

Num.

AICc

Weights

Likelihood

Par

Deviance

Φ(.) p(.) λ(.)

410.07

0.00

0.611

1.000

3

49.24

Φ(t) p(.) λ(.)

411.13

1.06

0.360

0.589

7

41.58

Φ(.) p(.) λ(t)

416.71

6.64

0.022

0.036

7

47.17

Φ(t) p(t) λ(.)

420.10

10.03

0.004

0.007

12

38.85

Φ(t) p(t) λ(t)

421.56

11.49

0.002

0.003

14

35.35

Φ(.) p(t) λ(t)

424.00

13.93

0.001

0.001

11

45.17

Table 2.8. Badger abundance estimates derived from both the POPAN model and
estimates of λ derived from Pradels-λ model. A similar trend in population growth of
badgers in the regional study area from 2003-2008 is shown using both methods.
Year

POPAN (SE)

Pradels-λ = 1.249

2003

12

-

-

2004

19

(0)

-

2005

38

(4.2)

38

2006

50

(4.7)

47

2007

56

(3.6)

59

2008

72

(3.5)

74

63

Figure 2.14. Number of underpass crossing structures monitored (n = 41) by size and
percentage of use by badgers in south-central BC, 2008 – 2009.

Figure 2.15. Number of underpass crossings documented by identified study animals,
known juveniles, and unknown badgers in south-central BC, 2008 – 2009.

64

around the entrance and leaving, or simply passing by the opening of the structure in the
right-of-way.
Study animals accounted for 51% of detected crossings while the other 49% consisted
of known juveniles and untagged or unknown badgers. Three badgers were identified as
study animals: an adult male (BM01) and two adult females (BF01 & BF07, Figure 2.16).
Both of the radio-tagged females were documented traveling through these structures with
their kit(s) in late summer (Figure 2.17).
Of the 40% of badgers that were classified as unknown individuals, nine percent were
identified as males based on sexual dimorphism features (i.e., head and body structure).
Facial-stripe recognition showed that repeated use of the structures was also happening by
badgers not in our tagged sample (Figure 2.18).
Surveys of 114 culverts in the study area where badgers were known to occur
revealed that 24% of these structures (n = 27, SD = 0.42) were physically impassable to
badgers. Common factors that limited the passage of badgers (and other wildlife) were
infilling with soil (37%), physical damage to the culvert (23%), standing water (23%),
blockage by debris (13%), and inaccessibility (hanging entrance, 4%).
The two compared segments of Highway 97 where badger road mortality occurred
had significantly different number of suitable crossing structures: the north segment had 45%
passable structures (9 of 20) while the south segment contained 88% passable structures (22
of 25; Fisher’s exact test p = 0.003). The north segment had 1.55 passable structures/km
whereas the south segment contained 2.62 passable structures/km. Overlaying known badger
road mortality sites revealed that a rate of 1.72 badger deaths/km along the north segment
were considerably higher than the 0.60 badger deaths/km from the south segment.
Discussion
Why are badgers susceptible to road mortality?
In my study area, a variety of temporal and spatial factors predispose badgers to the
risk of road mortality. Road mortality showed seasonal patterns that were consistent with
higher mobility periods such as rearing of young (May-July), dispersal (SeptemberNovember) and breeding season (July-August). In Spain, Grilo et al. (2009) found that the

65

Figure 2.16. Adult female (BF01) with male kit (right) exiting 500 mm culvert under
Highway 97 in south-central BC, 21 July 2009.

Figure 2.17. Facial pattern recognition was used to identify unique individual badgers
that passed through underpass structures in south-central BC, 2008 – 2009. Radiotagged female (left) and unknown male (right).

66

highest periods of road mortality for fox and marten were when feeding visits increased due
to the energetic demands of the litter. There appears to be similar evidence for badgers in my
study area, as two females were struck and killed on highways either in pursuit of prey, or
delivering prey to waiting kits, thus orphaning several dependent juveniles. Based on remote
camera data at natal dens, as juvenile rearing progressed beyond mid-May females traveled
extensively to capture and deliver prey to the den, followed by moving kits to a new maternal
den by mid-June. This behaviour increased both female and juvenile mortality risk as they
crossed major roads during this early summer period.
Juveniles dispersing from their natal territory across the landscape in the fall to
establish home ranges may be unfamiliar with the risk of roads. This study revealed that adult
females living in close proximity to roads were crossing under highways along with kits.
Later, the independent juveniles continued to use the underpass crossings, suggesting that
these individuals were learning or incorporating these structures as part of their regular
movement patterns, thereby reducing their road mortality risk. Road mortality risk may be
elevated for dispersing juveniles originating from a natal territory far from major roads as
they may be naïve to the risk of crossing high traffic-volume highways and be more
susceptible to vehicle collisions. Road mortality was less frequent in the fall and sex ratios of
badgers were equal suggesting that these road mortalities may be dispersing juveniles
searching for suitable habitats before the onset of winter.
Male badgers had considerably larger home ranges and traveled greater distance than
females, particularly during the breeding season from late June to early September, which
resulted in most males (4 of 5) being struck and killed on roads in the region. Males tripled
their mean movement rates and distances over females during summer, resulting in at least a
fivefold increase in home range size over females. Being a polygynous species, long distance
movements made by males likely occur as they are seeking receptive females during the
breeding season, suggesting females may be a limited resource, occurring sporadically in
patches across the landscape (Clutton-Brock and Harvey 1978, Sandell 1989).
Movement patterns also varied temporally, with animals documented crossing
highways from several times a day to once every several days, with most crossings occurring
during the summer season. Data from the animals outfitted with GPS dataloggers suggests
many of these crossing are not detected using conventional radio-telemetry methods, being

67

most frequent in late evening through to the early morning period. The GPS data from this
study indicate that badgers move more extensively during foraging and traveling periods in
summer than previously estimated (Hoodicoff 2003, Kinley and Newhouse 2008).
Distinct spatial patterns emerged in the badger road mortality data. Despite livecapturing three of five males > 9 km from a major highway, home ranges for all male
badgers in the study eventually spanned major transportation corridors. In contrast, the
smaller home ranges of females were often isolated from major roads. Thus, males appeared
more susceptible to road mortality across their range in the study area; this may in part be
attributed to both searching for receptive females during breeding season and seeking a
consistent prey source at specific times of the year, particularly in the fall when badgers are
accumulating fat reserves. Over the course of the study, females whose home ranges spanned
highways had lower survival rates than those females isolated from highways. The home
ranges of females spanning roadways consisted of mostly private land with a matrix of public
and private land between core areas [for definition of badger ‘core areas’ see (Hoodicoff et
al. 2009)]. In contrast, female home ranges that were distant from highways were primarily
composed of larger tracts of public land, had fewer roads, and consisted of a mixture of
native pocket grasslands adjacent to riparian corridors with open forests dominating the
uplands.
Other studies documenting carnivore road mortality recorded higher mortality in low
and intermediate traffic volume road segments, suggesting avoidance of high traffic volume
(Grilo et al. 2009), while others found high traffic volume had significant impacts on a
similar species, the Eurasian badger (Clarke et al. 1998). Clarke (1998) surmises that in
addition to causing direct mortality, major roads may also reduce dispersal and colonisation
movements. Of the two highways that occurred in the study area, the major roadway
(Highway 97) contributed to the majority of badger fatalities, while only a small percentage
occurred on the secondary roadway (Highway 24). It is unclear if traffic volume, amount of
suitable adjacent roadside habitat, or both acting together best explains the discrepancy
between badger fatalities along these highways.
I observed strong associations among highway location, home range orientation and
badger fatalities. These data highlighted three segments of Highway 97 in the study area
where the majority of mortalities occurred. Preferred badger habitat types of non-forested

68

upland with improved pasture/agricultural were consistent at these sites. Annual and summer
daily traffic volumes were three times higher on Highway 97 and it appears that this highway
bisects more suitable habitat, resulting in a peak of badger-vehicle collisions coinciding with
breeding season and high summer traffic volumes.
Is the population significantly impacted by road mortality?
The main ways that roads may affect wildlife populations are by (1) affecting the
survival rate through direct mortality, (2) changing population structure through
disproportionate road mortality of either sex causing biased sex ratios, and (3) changing
movement rates or patterns that ultimately affects population connectivity (Row et al. 2007).
Results from this study indicate that the population is stable to increasing despite several
animals being lost annually to direct mortality, sex ratios appear stable, and habitat
connectivity in the region is a concern with continuing highway improvement and
development projects.
In the radio-telemetry component of this study 50% of radio-tagged individuals
succumbed to road mortality. Survival rate estimates derived from DNA methods across the
region from the Burnham (Both) model suggest a highly variable estimate from year to year
for ‘captured’ badgers ranging from 45% to 78%, with an overall recovery rate of 28% of
marked dead individuals over the course of the study. The overall survival estimate of 75%
for all years and genders combined suggests that road mortality plays an important factor in
the long-term persistence of badgers in the region. Recovery rates for radio-tagged animals
and corrections for missed roadside deaths suggest that approximately 25% of the local
badger population annually succumbed to road mortality over the course of the study. Other
factors likely contribute to the overall mortality rate for badger populations in BC including
predation (Newhouse and Kinley 2001, Hoodicoff 2003) and persecution
(jeffersonii_Badger_Recovery_Team 2008). Despite no study animals dying from either of
these factors, there were two reported instances of shooting of untagged badgers in the local
study area (one confirmed mortality).
Attempts were made to equally sample badgers living near and distant from major
roads, and over half were captured a considerable distance from highways (> 4 km), but due
to their long-distance movements they still were exposed to highways in the region.
Regionally, areas in the western portion of the larger study area where we deployed hair-

69

snags had a lower density of secondary roads with no highways present and thus survival
would be expected to be much higher in those areas. It should be stressed that the increasing
population trend is region-wide with the majority of road mortality occurring within the local
study area where most of the study animals resided. Although age classes of dead animals
were not determined, the possibility of dispersing juveniles may account for the continued
but lower rates of road mortality outside of the breeding season. Road mortality can have
short-term consequences on population viability and it appears that badgers are most
vulnerable to increased traffic volumes during summer on highways that bisect high-quality
badger habitats. The reduction of traffic volume on major highways during the remainder of
the year provides a respite from direct mortality as badgers have restricted their movements,
either in anticipation of rearing young in the spring or preparation for the onset of winter. It is
unclear at what threshold that an increase in either road density or traffic volumes through
high-quality badger habitats may have on the long-term persistence of badgers in the region
but direct mortality will likely continue to exert a strong effect.
Population modeling from this study indicates that the regional population is stable or
increasing. The two most important assumptions for modeling population size and growth
rate are having a consistent study area size and equal catchability of animals (Cooch and
White 2000). While these assumptions were considered met in this study (particularly equal
catchability), achieving a consistent sampling area was likely confounded by increasing
efficiency and intensity of our DNA hair-snagging methods. Much of the population growth
early on in the project may be attributed to badgers being discovered in concert with better
sampling methods, thus increasing from a minimum number alive (12) to an estimate of 72
animals. Additionally, many of the recaptures we detected were likely dispersing juveniles
that provided population growth projections that would not be attributed to by immigration to
the study area. Therefore, areas where females occurred at the edge of the study area may not
have been sampled equally over the course of the project and dispersing juveniles from these
areas would act as a form of immigration. However, increasing regional badger population
estimates were also supported by anecdotal reports of landowners in the study area that the
badger population was on the increase during the course of the study from 2005 – 2009.
The effects of road mortality may explain the drastic sex bias in my sample of
captured badgers. My goal was to capture equal numbers of both sexes, but I captured twice

70

as many females than males (11 ♀: 5 ♂); this bias may have been attributed to my livecapture attempts being focused in areas with high quality badger habitat in summer resulting
in the capture of resident females rather than wide-ranging males. In addition, movement and
road mortality data would suggest that males were disproportionately being struck and killed
on roads before I could detect them. Despite this gender bias in my local study animals, the
data from the DNA hair-snagging sessions from 2003 – 2008 over a larger area of the region
suggests a more equal (50:50) sex ratio (48♂:52♀). Sex ratios of documented litters (n = 12
kits) using DNA methods were also close to 50:50 (58♂:42♀). Overall, results from the
long-term genetic sampling suggest that the population sex ratio in the region does not appear
to be affected by road mortality despite the high mortality rates within the local study area
where highways bisected badger habitat.
The apparent increase in badger abundance and distribution in the regional study area
could be in part due to a similar lag in population growth following recently improved
foraging conditions as seen elsewhere in the province (Kinley and Newhouse 2008).
Currently connectivity of badger metapopulations in the Cariboo region could be considered
favourable. In recent years, land clearing for agriculture on private lands combined with a
recent mountain pine beetle (Dendroctonus ponderosae) epidemic has created more open
landscapes dominated by graminoids and forbs (which support colonial rodent populations)
and improved range conditions favouring an abundant and diverse small mammal
community. Additionally, data from GPS-tagged animals reveal that badgers are crossing
secondary highways (Hwy 24) during periods of lowest traffic volumes (early morning,
night) resulting in less direct mortality thereby allowing movement of individuals across the
eastern portion of the study area. Despite suitable habitat conditions in the region and lower
road mortality risk along the lower traffic-volume secondary highway (Hwy 24), the ongoing
incidences of high road mortality along the Highway 97 corridor continues to be a concern
regarding the flow of individuals between metapopulations across the northern portion of the
region.
Mitigation and/or compensation
My results provide evidence that underpass structures are a feasible option for
reducing direct mortality of badgers from vehicle collisions. Although I identified three
segments of highway with high incidences of badger mortality, to predict these sites in other

71

areas without intensive study will be difficult. Instances where there is good habitat near the
road but low current road mortality can indicate particularly important locations for
mitigation to restore historically depressed populations (Eberhardt et al. 2013). Data from
culvert surveys and road mortalities suggests a higher incidence of badger deaths/km of
highway when suitable crossing structures were limited, while the camera data indicated
frequent use of culverts by badgers, including family units, independent juveniles, and tagged
and unknown males. The 500 mm culverts were the most common type along roadways and
badgers appeared to use these smaller structures most frequently, while > 2000 mm
underpass structures showed some degree of use, suggesting that badgers may use a broad
range of underpass structures when available. My ability to effectively monitor sufficient
numbers of structures simultaneously were limited by equipment and time constraints, and
further monitoring may reveal a wider range of structures used by badgers.
Smaller pipes or culverts are an increasingly common road mitigation tool for smallto medium-sized animals (Lankester et al. 1991, Forman et al. 2003). The recovery of the
Eurasian badger and other small animals has been greatly facilitated in parts of Europe
through the use of pipes, culverts, and other road underpass systems (Lankester et al. 1991).
In the Netherlands alone, over 300 such structures have been installed, contributing to the
remarkable recovery of the badger (Bekker and Canters 1997). In my study area, priority
should be given to maintenance of existing underpass structures (i.e., culverts) in areas that
are not slated for road improvements to optimize cost benefits by providing effective
structures currently in place for use by wildlife. Additionally, based on the results of this
study provincial government transportation and infrastructure planners are currently
incorporating wildlife- and badger-specific underpass crossing structures in highway
upgrades, including 4-laning along Highway 97 within known badger range.
Resident females residing on high-quality habitats that were considerable distances
[ ≥ 6 km based on radio-tagged females core home ranges (50%UD)] from highways and
major paved roads act as a source element in the population. These areas usually contained
older mature females that had a higher reproductive output (based on number of kits
emerging from the den) than younger females, or females occupying areas where ground
squirrels were absent. In addition to mitigating the direct effects of road mortality, particular
attention should also be focused on preserving or conserving the lands occupied by these

72

females. Acquisition of private parcels of land may be a feasible form of compensating for
habitat loss and direct mortality of badgers arising from the expansion of highway
infrastructure, while habitat enhancement and protection of high quality habitats on public
land should continue to be pursued by appropriate government agencies.
Remote motion cameras and GPS backpack methods proved useful for monitoring
badgers, albeit with some limitations (e.g., slow trigger speed of cameras missing animals).
For example, remote motion cameras successfully detected a GPS outfitted female (BF01)
passing through a culvert twice (Figure 2.19). Unfortunately, I could not compare her
detailed GPS movements at the time with the underpass camera data as the GPS battery unit
failed prematurely. Overall, using these methods in tandem revealed that specific highway
features (i.e., existing culverts) were being used by badgers, allowing me to document both
underpass crossing use and highway crossing rates that previously were underestimated.
Future advances in microGPS technology, in tandem with underpass monitoring (i.e., remote
cameras) remains a promising tool in detecting whether badgers utilize badger-specific
highway underpass structures or cross above grade in areas where movements are not
constrained (e.g., fencing).
Badgers in south-central BC appear to be susceptible to road mortality, but also are
exhibiting temporal and spatial variation in local population levels that may be attributed to
their existence at the northern periphery of their range. Few researchers have systematically
studied variation in population density over the geographic range of a species to determine
what environmental variables and population attributes may limit the abundance and
distribution of particular local populations (Brown 1984). Symes (2013) found that badgers
occurring at their northern range limit in BC were more active in winter than their
conspecifics further south, and that constraints imposed by the harsh winter environment may
limit the use of torpor as an energy-saving tactic. Rare carnivores living at the peripheral
edge of their range also are at greater risk to drastic fluctuation or stochastic events that
precipitate a sudden decline in population size (Lesica and Allendorf 1995). A
corresponding decline in genetic diversity may in turn impede adaptive responses to climate
change. Interestingly, projected climate change and associated effects of the mountain pine
beetle across central BC may improve badger habitat in the future by maintaining areas in an
open grassland/forest state with an abundant and diverse small mammal community.

73

Figure 2.18. Adult female badger (BF01) exiting culvert with GPS datalogger backpack
in south-central BC, 01 August 2008.

74

Conversely, these climatic changes may also have an impact on prey species distribution
including altering the synchronicity of food sources (timing of ground squirrel hibernation)
that may adversely affect foraging strategies for badgers in the region.
Using the “fine-filter” approach of protecting endangered species, subspecies or
populations is an important complement to more large-scale efforts (Franklin 1993).
Protecting biological diversity in the face of accelerating human-caused perturbations is a
major undertaking, and identifying and conserving habitats where badgers can fulfill their
life requisites (foraging, natal dens, over-winter sites) must be pursued not only to benefit
badgers but a wide range of species that are grassland and dry forest dependent.
Conclusion
Mitigation measures to reduce highway mortality coupled with habitat conservation
(to increase carrying capacity), appear key for allowing badgers to persist and perhaps
increase within south-central BC. This study represents the first attempt at investigating the
complex interaction of roads and their effects on North American badgers. Still, a number of
important issues concerning badger ecology and their relationship to roads remain to be
examined.
At the present time, the population appears to be stable or increasing with strong
recruitment compensating for road mortality losses in the region. Despite the positive trend in
population growth, densities are still a staggering low of one badger per 100 km2. Reported
densities further south in the United States indicate that badger populations can range from
one badger per 4 km2 up to one badger per 0.17 km2 (Warner and Ver Steeg 1995, Quinn et
al. 2006). Improvements in our sampling methodology may in part explain this increasing
population trend, but anecdotal observations also support that the population was expanding
in the region. Further DNA mark-recapture studies used to assess regional or
metapopulations should incorporate a rigorous systematic sampling design to reduce the bias
associated with the increasing efficiency in hair-snagging methods as developed throughout
the course of this study.
As mentioned, improved foraging conditions in tandem with public awareness may
have likely contributed to an increasing population that occurred during the course of this
study. It is unclear at what threshold increases in traffic volumes and expanding road
networks will render the local badger population unsustainable. It is unknown whether the

75

population will stabilize, decline, or follow a pattern of declines linked to direct mortality and
then increases associated with habitat conservation/improvements (increasing reproductive
output).
In the face of an expanding transportation system, population viability analysis (PVA)
for local badger populations should be explored. Road ecology studies similar to this study
have revealed through PVA that road mortality can increase extinction probability to
dangerous levels over the long-term (Row et al. 2007, Beaudry et al. 2008). Further data
collection on road mortality sites to help refine ‘hotspots’, as well addressing the relationship
between road mortality, prey abundance, and vegetation composition along right-of-ways
would certainly provide critical insights to assist the development of mitigation plans that can
minimize the adverse effects of roads on badgers.
Fencing in concert with under-passages has proven to be a successful method in
reducing badger-vehicle collisions in Europe (Broekhuizen and Derckx 1996, Vink et al.
2008). Clarke et al. (1998) and Lankester et al. (1991) conclude that road construction should
include fences to deter badgers (Meles meles) from crossing roads, and tunnels and
underpasses to increase survival. These methods could be employed in North America where
fencing trials could be conducted in areas that emerge as road mortality ‘hotspots’ with the
aim of significantly reducing badger deaths at these localized sites. Site-specific plans to
address the effects of constructing permanent drift-fencing as well as monitoring a range of
underpass structures should be pursued to determine other possible strategies for reductions
in badger-vehicle collisions in BC
LITERATURE CITED
Apps, C. D., N. J. Newhouse, and T. A. Kinley. 2002. Habitat associations of American
badgers in southeastern British Columbia. Canadian Journal of Zoology-Revue
Canadienne De Zoologie 80:1228-1239.
Beaudry, F., P. G. Demaynadier, and M. L. Hunter. 2008. Identifying road mortality threat at
multiple spatial scales for semi-aquatic turtles. Biological Conservation 141:25502563.
Bekker, H. G. J., and K. Canters. 1997. The continuing story of badgers and their tunnels.
Pages 344-353 in K. Canters, editor. Habitat fragmentation and infrastructure.
Ministry of Transport, Public Works and Water Management, Delft, Netherlands.
Belant, J. L. 2003. A hairsnare for forest carnivores. Wildlife Society Bulletin 31:482-485.

76

Bennett, A. F. 1991. Roads, roadsides, and wildlife conservation: a review. Pages 99-118 in
D. A. Saunders, and R. J. Hobbs, editor. Nature Conservation 2: the Role of
Corridors. Surrey Beatty and Sons, Chipping Norton, New South Wales, Australia.
British Columbia Ministry of Transportation and Infrastructure [MoTI]. 2009. Cariboo
Connector. in http://www.th.gov.bc.ca/cariboo_connector/.
_____. 2010. Traffic Data Program, Traffic Information GIS, AADT/SADT for Permanent
and Short Count Sites. in http://www.th.gov.bc.ca/trafficdata/.
Broekhuizen, S., and H. Derckx. 1996. Passages for badgers and their efficacy. Zeitschrift
Fur Jagdwissenschaft 42:134-142.
Brown, J. H. 1984. On the relationship bewtween abundance and distribution of species.
American Naturalist 124:255-279.
Burnham, K. P. 1993. A theory for combined analysis of ring recovery and recapture data.
Pages 199-213 in J. D. Lebreton, andP. North, editors. Marked individuals in bird
population studies. Birkhauser Verlag, Basel, Switzerland.
Burnham, K. P., and D. R. Anderson. 1998. Model selection and inference: a practical
information-theoretic approach. Springer-Verlag, New York, New York, USA.
_____. 2001. Kullback-Leibler information as a basis for strong inference in ecological
studies. Wildlife Research 28:111-119.
_____. 2004. Multimodel Inference - Understanding AIC and BIC in Model Selection.
Sociological Methods Research 33:261-304.
Clarke, G. P., P. C. L. White, and S. Harris. 1998. Effects of roads on badger Meles meles
populations in south-west England. Biological Conservation 86:117-124.
Clutton-Brock, T. H., and P. H. Harvey. 1978. Mammals, resources and reproductive
strategies. Nature 273:191-195.
Cooch, E., and G. C. White. 2000. Program MARK - a gentle introduction. 9th edition.
Available at http://www.phidot.org/software/mark/docs/book/.
Davies, J. M., T. J. Roper, and D. J. Shepherdson. 1987. Seasonal distribution of road kills in
the European badger (Meles meles). Journal of Zoology 211:525-529.
Dique, D. S., J. Thompson, H. J. Preece, G. C. Penfold, D. L. de Villiers, and R. S. Leslie.
2003. Koala mortality on roads in south-east Queensland: the koala speed-zone trial.
Wildlife Research 30:419-426.
Eberhardt, E., S. Mitchell, and L. Fahrig. 2013. Road kill hotspots do not effectively indicate
mitigation locations when past road kill has depressed populations. The Journal of
Wildlife Management 77:1353-1359.

77

Ferreras, P., J. J. Aldama, J. F. Beltran, and M. Delibes. 1992. Rates and causes of mortality
in a fragmented population of Iberian lynx Felis-pardina Temminck, 1824. Biological
Conservation 61:197-202.
Forman, R. T. T., J. Bissonette, A. Clevenger, C. Cutshall, V. Dale, L. Fahrig, C. Goldman,
K. Heanue, J. Jones, D. Sperling, F. Swanson, T. Turrentine, and T. Winter. 2003.
Road ecology : science and solutions. Island Press, Washington, DC.
Franklin, J. F. 1993. Preserving Biodiversity - Species, Ecosystems, or Landscapes.
Ecological Applications 3:202-205.
Goodrich, J. M., and S. W. Buskirk. 1998. Spacing and ecology of North American badgers
(Taxidea taxus) in a prairie-dog (Cynomys leucurus) complex. Journal of
Mammalogy 79:171-179.
Grilo, C., J. A. Bissonette, and M. Santos-Reis. 2009. Spatial-temporal patterns in
Mediterranean carnivore road casualties: Consequences for mitigation. Biological
Conservation 142:301-313.
Hoodicoff, C., and R. Packham. 2007. Cariboo Region Badger Project: Year End Report
2006-07. BC Ministry of Environment, 100 Mile House, BC. Report Technical, 38 pp
+ appendices.
Hoodicoff, C., S. 2003. Ecology of the badger (Taxidea taxus jeffersonii) in the Thompson
Region of British Columbia: Implications for Conservation. Thesis (M.Sc. Biology),
University of Victoria, August 2003.
Hoodicoff, C. S., K. W. Larsen, and R. D. Weir. 2009. Home range size and attributes for
badgers (Taxidea taxus jeffersonii) in south-central British Columbia, Canada.
American Midland Naturalist 162:305-317.
Hoodicoff, C. S., and R. Packham. 2006. Cariboo Region badger project: summary report,
2003-2005. Unpublished report to the BC Habitat Conservation Trust Fund. Summit
Environmental Consultants, Vernon, BC, and BC Ministry of Environment, 100 Mile
House.
Hovey, F. 1999. Home Ranger. Available from:
http://nhsbig.inhs.uiuc.edu/wes/home_range.html. in Kamloops, BC.
jeffersonii-Badger-Recovery-Team. 2008. Recovery Strategy for the Badger (Taxidea taxus)
in British Columbia. Prepared for the B.C. Ministry of Environment, Victoria, B.C.
45 pp.
Jones, M. E. 2000. Road upgrade, road mortality and remedial measures: impacts on a
population of eastern quolls and Tasmanian devils. Wildlife Research 27:289-296.
Kendall, K. C., J. B. Stetz, J. Boulanger, A. C. Macleod, D. Paetkau, and G. C. White. 2009.
Demography and Genetic Structure of a Recovering Grizzly Bear Population. The
Journal of Wildlife Management 73:3-16.

78

Kernohan, B. J., R. A. Gitzen, and J. J. Millspaugh. 2001. Analysis of animal space use and
movements. in J. M. Marzluff, editor. Radio Tracking and Animal Populations.
Academic Press,San Diego, CA.
King, T. L., and T. Burke. 1999. Gene conservation : identification and management of
genetic diversity. Blackwell Science, Oxford.
Kinley, T. A., and N. J. Newhouse. 2008. Ecology and translocation-aided recovery of an
endangered badger population. Journal of Wildlife Management 72:113-122.
Lalo, J. 1987. The problem of road kill. American Forests, American Forests. SeptemberOctober 1987.
Lankester, K., R. Vanapeldoorn, E. Meelis, and J. Verboom. 1991. Management perspectives
for populations of the Eurasian badger (Meles meles) in a fragmented landscape.
Journal of Applied Ecology 28:561-573.
Lesica, P., and F. W. Allendorf. 1995. When are peripheral populations valuable for
conservation. Conservation Biology 9:753-760.
Lindzey, F. G. 1978. Movement patterns of badgers in northwestern Utah. Journal of
Wildlife Management 42:418-422.
Messick, J. P. 1987. North American badger. Pages 586-597 in M. Novak, J. A. Baker, M. E.
Obbard, andB. Malloch, editors. Wild furbearer management and conservation in
North America. Ontario Trappers Association, and Ontario Ministry of Natural
Resources,Toronto, Canada.
Messick, J. P., and M. G. Hornocker. 1981. Ecology of the badger in southwestern Idaho.
Wildlife Monographs 76:1-53.
Millar, C. L., and K. A. Marshall. 1992. Allozyme variation of Port-Orford-cedar
(Chamaecyparis lawsonii): implications for genetic conservation. Forest Science
37:1060-1077.
Mulders, R., J. Boulanger, and D. Paetkau. 2007. Estimation of population size for
wolverines Gulo gulo at Daring Lake, Northwest Territories, Using DNA based
mark-recapture methods. Wildlife Biology 13:38-51.
Mumme, R. L., S. J. Schoech, G. W. Woolfenden, and J. W. Fitzpatrick. 2000. Life and death
in the fast lane: Demographic consequences of road mortality in the Florida ScrubJay. Conservation Biology 14:501-512.
Newhouse, N. J., and T. A. Kinley. 2001. Ecology of badgers near a range limit in British
Columbia. Technical report to Columbia Basin Fish & Wildlife Compensation
Program, Nelson, B.C.; and Parks Canada Agency, Radium Hot Springs, B.C.
Noss, R. F. 1987. From Plant-Communities to Landscapes in Conservation Inventories - a
Look at the Nature Conservancy (USA). Biological Conservation 41:11-37.

79

Paetkau. pers. comm. 2010. WGI project g0762, Cariboo Badger cover letter.
Paulson, N. J. 2007. Spatial and habitat ecology of North American badgers (Taxidea taxus)
in the native shrubsteppe ecosystem of Eastern Washington. Thesis (M.Sc. in natural
resource science), Washington State University, August 2007.
Pollock, K. H., J. D. Nichols, C. Brownie, and J. E. Hines. 1990. Statistical inference for
capture-recapture experiments. Wildlife Monographs 107:3-97.
Powell, R. A. 2000. Animal home ranges and territories and home range estimators. Pages
65-110 in L. Boitani, andT. K. Fuller, editors. Research Techniques in Animal
Ecology: Controversies and Consequences. Columbia University Press, New York.
Pradel, R. 1996. Utilization of capture-mark-recapture for the study of recruitment and
population growth rate. Biometrics 52:703-709.
Riley, S. P. D., J. P. Pollinger, R. M. Sauvajot, E. C. York, C. Bromley, T. K. Fuller, and R.
K. Wayne. 2006. A southern California freeway is a physical and social barrier to
gene flow in carnivores. Molecular Ecology 15:1733-1741.
Ripple, W. J., and R. L. Beschta. 2004a. Wolves and the ecology of fear. Bioscience 54:755766.
Rodgers, A. R., and A. P. Carr. 1998. HRE: The Home Range Extension for ArcView. in
Ontario Ministry of Natural Resources, Centre for Northern Forest Ecosystem
Research, Thunder Bay, Ontario, Canada.
Row, J. R., G. Blouin-Demers, and P. J. Weatherhead. 2007. Demographic effects of road
mortality in Black Ratsnakes (Elaphe obsolete). Biological Conservation 137:117124.
Ruediger, B. 1998. Rare carnivores and highways – moving into the 21st century. Pages 1016 in G. L. Evink, P. A. Garrett, D. Zeigler, and J. Berry, editor. Proceedings of the
International Conference on Wildlife Ecology and Transportation, February 10-12,
1998. Fort Myers, Florida. U.S. Department of Transportation, Federal Highway
Administration.
Sandell, M. 1989. The mating tactics and spacing patterns of solitary carnivores. Pages pp
164-182 in G. JL, editor. Carnivore behaviour, ecology, and evolution. Cornell
University Press, New York, USA.
Scheppers, T. L. J., T. J. Roper, A. C. Frantz, M. Schaul, E. e. Engel, P. Breyne, and L.
Schley. 2007. Estimating social group size of Eurasian badgers Meles meles by
genotyping remotely plucked single hairs. Wildlife Biology 13:195-207.
Schwarz, C. J., and A. N. Arnason. 1996. A general methodology for the analysis of openmodel capture recapture experiments. Biometrics 56:59-64.

80

Seaman, D. E., and R. A. Powell. 1996. An evaluation of the accuracy of kernel density
estimators for home range analysis. Ecology 77:2075-2085.
Smith, L. L., and J. C. K. Dodd. 2003. Wildlife mortality on U.S. Highway 441 across
Paynes Prairie, Alachua County, Florida. Florida Scientist 66:128-140.
Soulé, M. E., and L. S. Mills. 1998. Population Genetics: No Need to Isolate Genetics.
Science 282:1658-1659.
Steen, O. A., and R. A. Coupé. 1997. A field guide to forest site identification and
interpretation for the Cariboo Forest Region. BC Ministry of Forests. Victoria, BC.
Stoner, D. 1925. The toll of the automobile. Science 61:56-57.
Symes, S. A. 2013. Winter ecology of the North American badger (Taxidea taxus jeffersonii)
in the Cariboo region of British Columbia. MSc thesis, Thompson Rivers University,
Kamloops, BC, Canada.
Trombulak, S. C., and C. A. Frissell. 2000. Review of ecological effects of roads on
terrestrial and aquatic communities. Conservation Biology 14:18-30.
Vink, H. J., R. C. van Apeldoorn, and H. G. J. Bekker. 2008. Defragmentation measures and
the increase of a local European badger (Meles meles) population at Eindegooi, the
Netherlands. Lutra 51:75-86.
Weir, R. D., H. Davis, and C. Hoodicoff. 2003. Conservation strategies for North American
badgers in the Thompson and Okanagan regions. Habitat Conservation Trust Fund,
Artemis Wildlife Consultants, Armstrong, B.C. Report Final, pp 103.
White, G. C., and K. P. Burnham. 1999. Program MARK: survival estimation from
populations of marked animals. Bird Study 46 (Supplement):120-138.
Zielinski, W. J., F. V. Schlexer, K. L. Pilgrim, and M. K. Schwartz. 2006. The Efficacy of
Wire and Glue Hair Snares in Identifying Mesocarnivores. Wildlife Society Bulletin
34:1152-1161.

81

CHAPTER 3 – REDUCE MORTALITY AND INCREASE RECRUITMENT; A
STRATEGY FOR MAINTAINING BADGERS (TAXIDEA TAXUS) IN
SOUTH-CENTRAL BRITISH COLUMBIA
INTRODUCTION
In this chapter I present a focused discussion on how an understanding of the ecology
of badgers (particularly movements) may be integrated into habitat protection and restoration
plans, including mitigation for current and future road infrastructure projects, thereby
reducing impacts on this species. A brief overview of these findings followed by
management and mitigation recommendations is provided below. The reader may refer to
Chapter 2 for a more detailed description on methods and results from my study.
OVERVIEW OF STUDY FINDINGS
The badger is one of British Columbia’s (BC) most endangered mammals, with
estimates placing the total provincial population at fewer than 340 adults (jeffersonii Badger
Recovery Team 2008). Previous research projects in the province have isolated two main
factors contributing to the continued decline of badgers in BC: (1) a historical and current
decline in habitat, and (2) road mortality caused by major transportation corridors running
through valley bottoms where badgers exist. Realizing that badger mortality on roads in
south-central BC could threaten badgers, as it has elsewhere in the province, the BC Ministry
of Transportation and Infrastructure funded this research to study the interaction between
badgers and roads in the Cariboo region (see Figure 1.3, Chapter 1). Little to no literature
exists, to my knowledge, for research conducted specifically to understanding the effects of
roads on badgers in North America (Wright 1965, Case 1978). This study, focused on the
basic ecology and movements of the badger population, as a means to understand the factors
making these animals vulnerable to road mortality.
Mean home ranges for males were much larger than females, almost an eightfold
increase (♂ = 253.1 km2, ♀ = 29.4 km2). In addition to their extremely large home ranges,
males displayed extensive movements during summer, while reproductive females
maintained the smallest home ranges and restricted their movements during the kit rearing
period. Movements of adult females with kits were influenced by time of kit emergence from
the den, litter size, and period of time before natal den abandonment.

82

Movement rates were five times greater for GPS outfitted badgers versus
conventional radio-telemetry estimates [GPS 4,801 m/day (n = 5): VHF radio 840 m/day (n =
16)].
Both marked and unmarked badgers were recorded using a variety of structures, with
the majority of crossings occurring within the 500 mm diameter class. Data revealed that
both genders and family groups used culverts and livestock underpasses to pass under roads.
Both above- and below-grade road crossings were detected, with frequency varying by
individual, time of day, and seasonality.
Surveys of culverts in the study area where badgers occurred near roads revealed that
24% were physically impassable to badgers. Common factors that limited badger passage
were physical damage to the culvert, infilling with soil, blockage by debris, standing water,
and inaccessibility (overhanging opening).
Animals whose home ranges bisected highways were repeatedly observed crossing
highways, with half of the study animals succumbing to collisions with vehicles (8 of 16).
Due to their large home ranges, all five males in the study eventually crossed highway
corridors with four of five males dying on these major roadways. Sites of badger road
mortalities were spatially defined to identify areas of concentrated road mortality sites or
‘hotspots’ on major roadways in the south Cariboo.
Despite the high percentage of road mortality, long-term DNA data collection and
population modeling suggests an increasing or stable population of badgers in the regional
study area. At the present time, the high incidence of road mortality in this region seems to
be negated to some extent by strong recruitment. Increasing efficiency in sampling
techniques over the course of the study may have also contributed to the apparent population
increase in the region.
Home range and movement rate estimates for badgers in this study are some of the
largest reported in BC. Human-induced pressures on badger habitat often conflict with
seasonal movement patterns of badgers (i.e., breeding season) leading to habitat
fragmentation and direct mortality. The ability to understand these movement patterns and
provide mitigation measures (e.g., road underpasses, land conservation) that reduce direct

83

mortality and increase productivity will allow for the continued existence of this species near
their range limit in south-central BC.
MANAGEMENT AND MITIGATION
Management strategies for attempting to preserve populations of endangered
carnivores generally fluctuate between restoring habitat to allow increased recruitment or
reduce risks of mortality to increase survival rates (Ferreras et al. 2001). In this study,
protecting and restoring suitable habitat away from roads should result in more territories of
breeding females (i.e. increased carrying capacity). In theory, enhancing habitat (increasing
prey abundance) might reduce the average territory size, thus increasing the number of
breeding territories and reducing large movements during breeding and dispersal.
Conversely, knowing the mechanisms that predispose badger to road mortality should make
it possible to increase survival by instituting mitigating measures for an expanding
transportation sector. Below, I provide recommendations for both increasing recruitment
(habitat conservation/restoration) and minimizing direct (road) mortality of badgers in the
study area.
Habitat protection / restoration (increase recruitment):
Radio-tracking adult badgers revealed specific areas that contained important habitat
elements (suitable soils, prey) on public lands where suitable badger habitat was
limited (i.e., rocky soils). Land use management actions on public lands to improve
badger populations include creating Wildlife Habitat Areas (WHAs) that permit
limited development provided it does not harm or degrade habitats in specific areas.
At the present time, 3,819 hectares of quality habitats has been included for WHAs in
the Cariboo region of BC providing some degree of protection for these important
habitat features.
Natal dens of mature female badgers were often located on private lands considered
“high quality” habitats in the local study area. Dispersing juveniles from these land
parcels provide migrants for vacant territories. Land acquisition, including covenants
and easements by conservation organizations (i.e., The Nature Trust of BC) could be
a cost-effective mitigative solution for highway improvement projects. These
acquired parcels could serve to aid the recovery of badgers and other endangered

84

wildlife species, and preserve essential habitat connectivity. Habitat enhancement and
restoration efforts could be conducted on acquired conservation lands to support
adaptation and species resiliency for this peripheral population.
Radio- and GPS-tracking of badgers revealed the use of timber-harvested sites that
were in early seral stages. Specific forest harvesting and silvicultural techniques that
enhance habitat, in concert with continued grassland ecosystem restoration efforts
that promote an abundance of prey species (e.g., vole spp., ground squirrels) should
be pursued in areas where badgers are known to occur. Grassland restoration efforts
have been conducted for 2,900 hectares on public land in the Cariboo region of BC
from 2004 – 2010.
Landowners were more receptive to tolerating badgers on their properties when they
were made aware that the major prey source for badgers are those rodent species
generally considered ‘pests’ by the agricultural industry (see public brochure Appendix D). In order to counter the possible extirpation of badgers in the region,
stewardship by landowners of high-quality habitats occurring on private land should
be attempted. This management action would then continue to provide a source
population to fill the vacancies created by road mortality losses. Stewardship
activities targeted at educating and promoting badgers on landowner’s properties
should be continued.
Mortality reduction (highway specific):
This study found evidence revealing that badgers frequently used existing dry culverts to
pass under roadways in the Cariboo region of south-central BC. Previous projects reported
that existing underpass structures (i.e., culverts) may be used by badgers to cross under major
transportation corridors (Newhouse and Kinley 2001). Identifying areas of high road
mortality, seasonality of road mortality, adjacent vegetative cover modifications, roadside
topography (habitat), and types of crossing structures present all contribute to the knowledge
required for managers to implement mitigation measures to counter the effects of road
mortality of badgers. Recommendations from this study are occurring through the installation
of a variety of wildlife-specific underpass crossing structures, including badger-specific
underpasses along Highway 97 (Figure 3.1). Recommendations pertaining to the reduction of
badger road mortality in the region include:

85

Providing dry culverts as underpasses is a feasible form of mitigation for badgers
where road mortality occurs. Over a third of culverts monitored showed badgers (both
genders, family units, adults, and juveniles) crossing under highways in the study
area. Evidence suggesting that adult females are passing this knowledge on to
juveniles was documented by remote motion cameras.
Badgers crossing under highways were most frequently documented using between
500 – 800 mm corrugated metal pipe culverts with some use of larger structures, up to
2500 mm. It appears badgers will readily use these structures based on availability
and providing a range of size in structures will not only benefit badgers but a variety
of wildlife species.
Regularly conducted maintenance of culverts would be a cost-effective way of
providing underpass structures for wildlife. Approximately 24% of all culverts
surveyed in this study were considered impassable to badgers. These works include
repairing crushed ends, removing blockages, ensuring soil is removed from the
entrance for proper drainage, and constructing earthen ramps to overhanging culverts
to allow access for the short-legged badger.
A slight grade in the structure would prevent standing water buildup during regular
rainfall events. In light of badgers crossing under roads more frequently in the
summer, some seasonal flooding or extreme weather events resulting in excess runoff
would be acceptable short term impacts.
Earth substrate bottoms were a common feature for culverts frequently used by
badgers (Figure 3.2). There is evidence to suggest that badgers have particularly
sensitive feet (to detect burrowing rodents) and foreign metal surfaces may be
avoided.

86

Figure 3.1. Recently installed badger-specific underpass crossing structure along Highway 97 south of 100 Mile House BC, Canada,
2010.

87

Figure 3.2. Suitable existing underpass structure (500 mm corrugated pipe) with earth substrate bottom used by badgers to pass
under Highway 97 north of 100 Mile House BC, Canada, 19 August 2008.

88

Attempt to have more than two underpass structures per linear kilometer of highway.
Two segments of highway where radio-tagged badgers and badger-vehicle collisions
frequently occurred were assessed for suitable existing underpass structures. The
south segment had the lowest road mortality rates where 88% of existing culverts
were available for badger underpass use. Higher road mortality rates occurred on the
north segment where only 45% of existing culverts were available for badger
underpass use. A suitable crossing structure may not necessarily be a badger- or
wildlife-specific structure but an existing dry culvert.
Include access roads in wildlife underpass structure planning and design. Monitoring
revealed that access road culverts were used by badgers to travel along highway rightof-ways. Access roads were common at road mortality sites suggesting that animals
wandered onto the highway roadbed when encountering a blocked passageway under
these accessory roads. A common feature at road mortality and above grade crossing
sites were raised banks adjacent to the highway. Consider installation of wildlifespecific culverts along stretches where there is a raised bank on one or both sides of
right-of-way, especially in areas where pasture/agricultural land is adjacent to the
roadway (Figures 3.3A and B). These sites were not considered for drainage purposes
so there are minimal existing culverts at these sites.
Mowing of roadside vegetation during summer months when vegetation exceeds 30
cm should be conducted. Badgers were observed traveling along the shoulder of the
highway where tall dense vegetation can be used as cover when threatened by
approaching vehicles. By mowing and removing this tall grass-structure two to four
meters from the shoulder of the roadbed the animals would then be more exposed and
less likely to travel along the edge of the road.
Alternative vegetation should be explored for re-seeding large areas of right-of-ways
by replacing the tall (> 50 cm) herbaceous vegetation with lesser-palatable lower
growing species (< 30 cm). This can still achieve the erosion, soil conditioning, and
noxious weed reduction objectives of seeding right-of-ways with herbaceous
vegetation. Avoid species such as alfalfa (Medicago spp.) and clover (Trifolium spp.)
and other domestic graminoid species. The seed crop from these highly palatable
forage species cause an abundant small mammal community to flourish along the

89

600 mm corrugated
metal culvert

Figures 3.3A and B. Examples of raised banks adjacent to highway corridor. Badger
road mortality site (top) where no crossing structures were present nearby (> 250 m)
and successful crossing site used repeatedly by badger (below). Both sites along
Highway 97 north of 100 Mile House BC, Canada.

90

right-of-way which attracts badgers to forage along roadsides. An additional benefit
would be the reduction in tall dense cover that badgers prefer when traveling along
highway right-of-ways during daylight hours, while also reducing the attraction for
large herbivores (i.e., deer) to the roadside.
Pre- and post-monitoring of structures (e.g., cameras, track plates) should be
implemented to measure success or failures of wildlife underpass crossing projects. It
is often noted that there is a need for additional evidence-based support that wildlife
crossing structures contribute to the long-term persistence of wildlife populations and
long-term monitoring is recommended (Clevenger and Sawaya 2010).
Future study
Much work is still needed to expand on the understanding of the complex relationship
between badgers and roads from this study. As with most exploratory projects, a number of
important questions arise to be answered. Recommendations to improve the knowledge of
road ecology of badgers are as follows:
Initiate culvert/underpass studies to expand on the knowledge of preferred structure
size, density of suitable structures, and placement on the landscape for not just
badgers, but other species that may benefit from increasing permeability across
roadways.
Conduct a comprehensive Population Viability Analysis (PVA) for badgers to
determine long-term risk assessment for metapopulations in the region.
Long-term, consistently-collected road mortality data would aid immensely in
refining the spatial and temporal patterns of mortality ‘hotspots’ and should be
continued by appropriate government agencies.
Pursue the development of movement models for badgers incorporating movement
data, habitat, soils, and landscape use, which may identify important corridors and aid
in resolving current and future roadway conflicts. This should include projected
habitat alterations induced by potential climate change effects, an expanding
transportation infrastructure, and human development.
Conduct fencing trials in areas that emerge as road mortality ‘hotspots’ with the aim
of significantly reducing badger deaths at these localized sites.

91

Vegetation trials along highways should be conducted to determine if amount of
cover influences badger attraction/avoidance and species richness/diversity has an
effect on prey abundance and distribution.
Conduct periodic systematic non-invasive DNA capture-recapture sampling of
badgers every three to five years to assess local population trends and status.
Identify ecological factors that contribute to colonial fossorial rodent (i.e., Columbian
ground squirrels, yellow-bellied marmots) colonization and distribution in BC. It has
been postulated that badger distribution in BC is linked to the distribution of these
prey species (jeffersonii Badger Recovery Team 2008).
DISCUSSION
My study supports the notion that badger conservation in the region may be best
served by a two-pronged approach aimed at reducing mortality coupled with increasing
recruitment via habitat protection and restoration. Over the last decade, natural resource
managers and transportation agencies have become increasingly aware of the effects of roads
on wildlife and remedial measures to counter these effects are an emerging science (Ruediger
1998, Clevenger and Sawaya 2010). Wildlife crossing structures are increasingly used by
transportation agencies to meet the need to allow animals to cross roads with reduced hazards
to both motorists and wildlife (Goosem et al. 2001, Mata et al. 2005). Studies have
demonstrated that a broad range of species will use wildlife crossing structures (Ng et al.
2004, Mata et al. 2005), which can reduce road-related mortality for some species (Clevenger
et al. 2003). Despite reducing direct mortality of carnivores, it has been stressed that the
usefulness of preserving areas that act as refugia for source populations will also have a large
affect on the population persistence (Ferreras et al. 2001)
Results from population modeling for this study appear somewhat contradictory:
direct mortality from collisions with vehicles seems overwhelming, but yet DNA capturerecapture data suggests a stable or increasing population. At the present time, the high
incidence of road mortality in the study area seems to be negated to some extent by strong
recruitment. It should be stressed that the population trend is region-wide with the majority
of road mortality occurring within the local study area. Bias from an increasing efficiency in
DNA sampling methods over several years also likely occurred.

92

In recent years land clearing for cultivation/agriculture on private lands combined
with the recent mountain pine beetle epidemic has favoured the expansion of open
landscapes across the region dominated by graminoids and forbs that are favourable to
colonization of rodents. The apparent increase in badger distribution could be in part due to a
similar lag in population growth following recently improved foraging conditions, as seen
elsewhere in the province (Kinley and Newhouse 2008). Another factor contributing to this
apparent increase may be in part to the social impacts emanating from this study. Fieldwork
was frequently carried out on private lands resulting in an elevated awareness and message
that badgers occurring on these lands were providing beneficial ecosystem services (i.e.,
rodent control) resulting in less human persecution. Certainly, efforts should be maintained to
keep the public aware of ecosystem benefits of badgers residing on both private and public
lands.
Badgers appear to be a favourable candidate in the wake of possible climate change
as peripheral populations often occur in atypical habitats where adaptation and resiliency to
this change may be reflected along the northern limits of their range (Root and Schneider
2006). Due to badgers existing at the northern periphery of their range, sudden stochastic
events, possibly due to the effects of climate change [e.g., prey collapse, extreme weather
(Walther et al. 2002, Thomas et al. 2004)], coupled with increased mortality due to an
expanding transportation sector may lead to a drastic decline in the local population
eventually leading to extirpation as observed in the early 90’s in the south-east corner of the
province (Newhouse and Kinley 2001). Despite the apparent population increase of badgers
in the region, little is known on the distribution and abundance of prey species. Many
landowners noted that populations of colonial rodent populations in the early part of the
2000s were at an all-time high, having declined dramatically since the recent influx of
badgers. Therefore, the best way to assure the survival of badgers is to continue to provide
refuge habitats with high carrying capacity coupled with the reduction of direct (road)
mortality so that populations can adapt to changes in the environment.
A multi-scale approach to mitigation planning, including land acquisitions that extend
beyond the local study area, can ensure that mitigation optimally adds to regional ecological
plans (Thorne et al. 2009). The wide range of underpass use by a number of species indicate
that species-specific considerations in tandem with broader ecosystem planning may

93

facilitate several different underpass types at varying frequencies to be incorporated into
transportation projects (Mata et al. 2005), thus reducing mortality and increasing connectivity
for not just badgers but a suite of wildlife species. Infrastructure agencies are not necessarily
able to meet all the conservation goals for a region through their mitigation obligations, and
conservation planners should consider the opportunities of partnering with these groups as
society strives to achieve regional sustainability (Thorne et al. 2009).
The data collected in this study provide an initial understanding of badger road
ecology in the Cariboo region of south-central BC, in particular how roads interact with and
shape the animal’s movements, distribution, and demographics. It is unclear at what
threshold that an increase in either road density or traffic volumes through high-quality
badger habitats may have on the long-term persistence of badgers in the region but direct
mortality will likely continue to exert a strong effect. My goal is that these findings, in
tandem with habitat conservation and restoration efforts provide guidance for managers,
biologists, and engineers to implement effective management plans for this unique carnivore.
I believe that such efforts will increase this species’ resilience and allow adaptation in the
face of environmental change. The continued existence of this ecologically important
mesocarnivore will allow people the opportunity to discover these fascinating creatures that
are contributing to healthy ecosystems across southern BC.
LITERATURE CITED
Case, R. M. 1978. Interstate highway road-killed animals: a data source for biologists.
Wildlife Society Bulletin 6:8-13.
Clevenger, A. P., B. Chruszczc, and K. E. Gunson. 2003. Spatial patterns and factors
influencing small vertebrate fauna road-kill aggregations. Biological Conservation
109:15-26.
Clevenger, A. P., and M. A. Sawaya. 2010. Piloting a non-invasive genetic sampling method
for evaluating population-level benefits of wildlife crossing structures. Ecology and
Society 15.
Ferreras, P., P. Gaona, F. Palomares, and M. Delibes. 2001. Restore habitat or reduce
mortality? Implications from a population viability analysis of the Iberian lynx.
Animal Conservation 4:265-274.
Goosem, M., Y. Izumi, and S. Turton. 2001. Efforts to restore habitat connectivity for an
upland tropical rainforest fauna: a trial of underpasses below roads. Ecological
Management & Restoration 2:196-202.

94

jeffersonii-Badger-Recovery-Team. 2008. Recovery Strategy for the Badger (Taxidea taxus)
in British Columbia. Prepared for the B.C. Ministry of Environment, Victoria, B.C.
45 pp.
Kinley, T. A., and N. J. Newhouse. 2008. Ecology and translocation-aided recovery of an
endangered badger population. Journal of Wildlife Management 72:113-122.
Mata, C., I. Hervas, J. Herranz, F. Suarez, and J. E. Malo. 2005. Complementary use by
vertebrates of crossing structures along a fenced Spanish motorway. Biological
Conservation 124:397-405.
Newhouse, N.J., and T.A., Kinley. 2001. Ecology of badgers near a range limit in British
Columbia. Technical report to Columbia Basin Fish & Wildlife Compensation
Program, Nelson, B.C.; and Parks Canada Agency, Radium Hot Springs, B.C.
Ng, S. J., J. W. Dole, R. M. Sauvajot, S. P. D. Riley, and T. J. Valone. 2004. Use of highway
undercrossings by wildlife in southern California. Biological Conservation 115:499507.
Root, T. L., and S. H. Schneider. 2006. Conservation and climate change: The challenges
ahead. Conservation Biology 20:706-708.
Ruediger, B. 1998. Rare carnivores and highways – moving into the 21st century. Pages 1016 in G. L. Evink, P. A. Garrett, D. Zeigler, and J. Berry, editor. Proceedings of the
International Conference on Wildlife Ecology and Transportation, February 10-12,
1998. Fort Myers, Florida. U.S. Department of Transportation, Federal Highway
Administration.
Thomas, C. D., A. Cameron, R. E. Green, M. Bakkenes, L. J. Beaumont, Y. C. Collingham,
B. F. N. Erasmus, M. F. de Siqueira, A. Grainger, L. Hannah, L. Hughes, B. Huntley,
A. S. van Jaarsveld, G. F. Midgley, L. Miles, M. A. Ortega-Huerta, A. T. Peterson, O.
L. Phillips, and S. E. Williams. 2004. Extinction risk from climate change. Nature
427:145-148.
Thorne, J. H., P. R. Huber, E. H. Girvetz, J. Quinn, and M. C. McCoy. 2009. Integration of
Regional Mitigation Assessment and Conservation Planning. Ecology and Society 14.
Walther, G. R., E. Post, P. Convey, A. Menzel, C. Parmesan, T. J. C. Beebee, J. M.
Fromentin, O. Hoegh-Guldberg, and F. Bairlein. 2002. Ecological responses to recent
climate change. Nature 416:389-395.
Wright, P. L. 1965. Observations on reproductive cycle of American badger (Taxidea taxus).
Journal of Reproduction and Fertility 9:364-367.

95

APPENDIX A. Reproductive parameters for female badger natal dens monitored
during the study in south-central BC, Canada 2008-2010.
Badger

# of

ID ♀

Year

kits

Kit emergence

Leave natal den

# of days

BF05

2008

2

14/05/2008

11/06/2008

28

BF03

2008

2

20/05/2008

07/06/2008

18

BF11

2009

2

28/05/2009

14/06/2009

17

BF07

2009

2

01/06/2009

24/06/2009

23

BF05

2009

3

23/05/2009

03/06/2009

11

BF03

2009

4

23/05/2009

29/05/2009

6

BF02

2009

3

25/05/2009

30/05/2009

5

BF01

2009

1

25/05/2009

31/05/2009

6

BF10

2010

2

01/06/2010

01/06/2010

0

BF11

2010

3

08/05/2010

29/05/2010

21

BF09

2010

2

09/05/2010

29/05/2010

20

BF08

2010

2

06/05/2010

01/06/2010

26

BF03

2010

3

14/05/2010

05/06/2010

22

96

APPENDIX B. Fates of radio-tagged badger study animals, capture dates, known
mortality dates, fate, and number of days radio-monitored as of June 2010.
Badger ID

Sex

Capt. date

Mort. date

Fate

Days monitored

BF01

♀

2007-07-29

?

unknown

842

BF02

♀

2007-07-29

2010-05-17

roadkill

1023

BF03

♀

2007-08-02

-

alive

914

BF04

♀

2007-08-05

2008-09-29

roadkill

421

BM01

♂

2007-08-05

2009-07-22

roadkill

717

BF05

♀

2007-08-15

-

alive

815

BM02

♂

2008-05-19

-

alive

415

BM03

♂

2008-05-21

2008-08-18

roadkill

89

BM04

♂

2008-05-23

2008-07-13

roadkill

51

BM05

♂

2008-06-18

2008-08-30

roadkill

73

BF06

♀

2008-08-22

2009-05-23

roadkill

274

BF07

♀

2008-08-24

2009-07-02

roadkill

312

BF08

♀

2009-05-25

-

alive

191

BF09

♀

2009-05-25

-

alive

191

BF10

♀

2009-05-27

-

alive

250

BF11

♀

2009-05-28

-

alive

249

97

APPENDIX C. Poster for solicitations for badger sightings from the public.

98

APPENDIX D. Public brochure handout (front and back) distributed in the south Cariboo region, 2007-2010.

99