EFFECTIVENESS OF ROAD MORTALITY MITIGATION IN A NORTHERN
COMMUNITY OF SNAKES
by

JADE AMBER SPRUYT
BSc, University of Calgary, 2018

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 2024

Thesis Examining Committee:
Karl W. Larsen (PhD), Thesis Supervisor
Professor, Department of Natural Resource Sciences, Thompson Rivers University
Christine Bishop (PhD), Thesis Co-Supervisor
Research Scientist Emeritus, Environment & Climate Change Canada
Adjunct Faculty, Thompson Rivers University
Laura Lamb (PhD), Committee Member
Professor, Department of Economics, Thompson Rivers University
Pamela Rutherford (PhD), External Examiner
Associate Professor, Department of Biology, Brandon University

i

ii
ABSTRACT
Roads and vehicular traffic are one of the most wide-spread threats to wildlife
globally. They are particularly detrimental to long-lived species with lower population
growth rates that are less able to naturally compensate for increases in anthropogenic
mortality. Road crossing structures and fencing often are implemented to mitigate the
impacts of road mortality on wildlife. However, follow-up studies assessing their efficacy are
limited, especially for reptiles who are particularly vulnerable to road effects. Although
numerical (population) responses to mitigation structures are strong indicators of the overall
effectiveness of the mitigation structures, they require intensive and/or long-term data.
However, functional (behavioural) responses also may provide good indicators of the
response to mitigative efforts; thus, functional and numerical responses used in tandem may
provide a more robust assessment of the impact that the mitigation structures, including those
designed to reduce road impacts, are having on populations.
I conducted a detailed assessment on the short-term, immediate impact of recentlyinstalled ecopassages and directional fencing on a threatened Western Rattlesnake (Crotalus
oreganus) population south-central British Columbia, Canada. Using road surveys, traffic
monitoring, and mark-recapture methods, I analyzed trends in roadkill rates and population
size during a 6-year period (2015-2020) that encompassed the periods before, during, and
immediately after (two years) the installment of ecopassages under a highway. Roadkill rates
appeared to decrease after mitigation installation (0.06 ± 0.03 SE deaths/km/day before, 0.03
± 0.01 SE deaths/km/day after) despite an increase in traffic (302 vehicles/day before, 454
vehicles/day after), yet population trends for adults did not indicate a clear trajectory towards
recovery.
Wildlife cameras were installed in the ecopassages, and I used photograph data to
quantify spatial and temporal usage patterns and assess the immediate post-installation
effectiveness of drift fencing for the Western Rattlesnake (Crotalus oreganus) and two other
threatened species in the same community: the Great Basin Gophersnake (Pituophis catenifer
deserticola), and Western Yellow-bellied Racer (Coluber constrictor mormon). I quantified
“appearances” (any time an individual appeared on camera), and “passages” (where a snake

iii
was documented travelling through an ecopassage within a 30 minute time span). Across two
years (2019 and 2020) appearances and passages were, respectively, 115 and 12 for
rattlesnakes, 183 and 29 for gophersnakes, and 3061 and 748 for racers. Although racers
clearly used the ecopassages significantly more than the other two species, both appearances
and passages increased for rattlesnakes, gophersnakes, and racers from 2019 to 2020. The
proportion of entrances with fences showed increased usage from 2019 to 2020, however
these changes only were significant for rattlesnakes. Temporal and spatial use patterns
differed among species, and I postulate that this was due to differences in movement patterns
and habitat preferences.
This study highlights the short-term yet complex response of snake communities to
the effects of roadway mitigation, as the animals presumably encounter, adjust, and respond
to the new structures in their environment. Despite the three snake species inhabiting the
same environment, different responses were demonstrated towards the new mitigation
structures. Short-term assessments of the response to mitigation efforts, such as applied in
this study, likely illustrate a ‘shock phase’ in the wildlife community, and should be coupled
with longer term monitoring to gauge the full effect of the conservation actions.
Keywords: road ecology, road mortality, mitigation, ecopassages, drift fencing, population
estimate, camera monitoring, conservation, reptile, snake

iv
ACKNOWLEDGEMENTS
Firstly, I would like to acknowledge that this research was conducted on the
traditional unceded territory of the Sylix (Okanagan) people. I feel honoured to have had the
opportunity to learn from this beautiful and fascinating land. I would also like to thank the
National Research Council Canada, Dominion Radio Astrophysical Observatory, The Nature
Trust of British Columbia, and the Clifton Ranch for allowing me to access these lands.
I would like to express my deepest gratitude to my supervisor Karl Larsen, for not
only providing endless encouragement and guidance, but also for introducing me to Linda
Ronstadt and the Eagles. I am also extremely grateful to Christine Bishop for her support
throughout the project, and her belief in my abilities as a field biologist. I also would like to
thank Stephanie Winton, for entrusting me with her project, teaching me everything I know
about snakes, and being infinitely patient while answering my non-stop stream of questions.
I am so grateful to all the field assistants and volunteers who contributed to this
project so enthusiastically. Specifically, thanks to Jake Veenman, Michelle Matson, and
Lauren Matus for enduring early morning road surveys and the quirks of the White Lake
Snake House. This research would not be possible without the other field crew members who
collected data throughout the seven-year span of this project: Jenny Petersen, Hannah
Butcher, Jennifer Barden, Anna Skurikhina, Cole Hooper, Vanessa Robinson, and Matt
Macpherson. Thanks to Dana Eye and Chloe Howarth who made den surveys so much fun
and writing my thesis bearable, and to my other lab-mates and classmates who made my
Masters experience so great - specifically Wendy Margetts and Marcus Atkins for providing
motivation via sweet treats.
I would also like to thank my family and close friends for their constant support and
trying their best to get excited about snakes. Your encouragement means the world.
This research was made possible with the support of many partners, including the BC
Ministries of Water, Land & Resource Stewardship (WLRS) and Transportation &
Infrastructure (MoTI). I would like to thank Jamie Leathem for advocating for this project,
and Donna Olsen and others at MoTI who helped get the ecopassages installed. I also want to

v
acknowledge Mike Sarrell for installing the drift fencing and for being a pioneer in the battle
against snake roadkill at White Lake.
Financial support for this project was provided by Interdepartmental Recovery Funds,
Environment and Climate Change Canada, BC Ministry of Water, Land & Resource
Stewardship, and TRU’s UREAP scholarship. All snakes were handled under the following
permits: Environment Canada Species at Risk Act (SARA-PYR-2015-0316, SARA-PYR2018-0427, SARA-PYR-2021-0587), British Columbia Wildlife Act (PE15-171661,
MRPE20-609553), and Thompson Rivers University Animal Use Protocol (102042,
101546).

vi
TABLE OF CONTENTS
Abstract………………………………………………………………………………..………ii
Acknowledgements…………………………………………………………………………...iv
Table of Contents……………………………………………………………………………..vi
List of Figures……………………………………………………………………………….viii
List of Tables………………………………………………………………………………….x
Chapter 1. Introduction……………………………………………………………………12
Wildlife Mortality Factors…………………………………………………………...12
Road Mortality and Mitigation………………………………………………………12
Snakes and Roads……………………………………………………………………13
Study Site…………………………………………………………………………….14
Research Objectives……………………………………………………………...…..15
Site Description……...……………………………………………………………….16
Literature Cited………………………………………………………………………22
Chapter 2. Transitional Response of Western Rattlesnakes (Crotalus oreganus) to the
Installation of Road Ecopassages and Drift Fencing………………………...………..25
Introduction…………………………………………………………………………..25
Methods………………………………………………………………………………27
Study Site…………………………………………………………………….27
Calculating Road Mortality Rates……………………………………………29
Monitoring Population Trajectory…………………………………………...30
Results………………………………………………………………………………..30
Road Surveys……………………………………………………………...…30
Traffic Rates………………………………………………………………….31
Roadkill Rates……………………………………………………………… 31
Mark-Recapture……………………………………………………………...31
Population Estimate………………………………………………………….31

vii
Discussion……………………………………………………………………………32
Literature Cited………………………………………………………………………39
Chapter 3. Different Snake Species Show Varying Responses To Ecopassage
Installation………..………………………………………………………………...……43
Introduction…………………………………………………………………………..43
Methods………………………………………………………………………………45
Study Site…………………………………………………………………….45
Camera Monitoring…………………………………………………………..47
Data Analysis……..………………………………………………………….59
Results………………………………………………………………………………..59
Discussion……………………………………………………………………………50
Literature Cited………………………………………………………………………59
Chapter 4. Conclusion……………………………………………………………………...64
Management Implications……………………………………………………………65
Future Research Considerations.…………………………………………………….67
Conclusion…………………………………………………………………………...68
Literature Cited…………………………………………………………………...….70
Appendix A. Jolly-Seber Estimates for Rattlesnakes in the White Lake Basin……………..72
Appendix B. Snake Road Mortality in the White Lake Basin……………………………….74
Appendix C. Snake Ecopassage Use in the White Lake Basin………………………………78
Appendix D. Ecopassage Use by Non-target Species in the White Lake Basin……………..79

viii
LIST OF FIGURES

Figure 1.1. Study site location within the Western Rattlesnake (Crotalus oreganus) range in
western North America (from Winton 2018)………………………………………...17
Figure 1.2. A) Mean monthly temperature (°C) and B) total monthly precipitation (mm) in
the South Okanagan during the study years (2015-2021) as measured at the Penticton
Regional Airport, British Columbia, Canada (49°N, 119°W; Environment and
Climate Change Canada 2024)………………………………………………………19
Figure 1.3. Timeline of snake research in the White Lake Basin, British Columbia………..20
Figure 1.4. Examples of snake road mortality mitigation measures in the White Lake Basin,
British Columbia, including A) a modified culvert equipped with a camera, with the
entrance unaccompanied by fencing, B) a newly-installed semi-circular ecopassage
equipped with a camera and fencing surrounding the entrance, and C & D) drift
fencing positioned to funnel snakes into under-road ecopassages………..………….21
Figure 2.1. Calculated Western Rattlesnake (Crotalus oreganus) roadkill rates (± 1 SE) (left
ordinate, black data points) with a linear regression line (y = -0.006x + 0.0613, R2 =
0.42, P = 0.17), and average traffic rates (right ordinate, bar graph) from 2015-2020.
Traffic data were unavailable for 2018. Solid arrow on bottom indicates when
ecopassages were installed in 2017; broken arrow indicates the installation of drift
fencing prior to the 2019 active season.…………..………………….………………34
Figure 2.2. A & B) Proportion of Western Rattlesnakes captured across size classes at six
focal dens in the White Lake Basin before (A) and after (B) ecopassages were
installed. Snakes with measured lengths falling exactly on an interval cut-off were
included in the larger category. C & D) Number of roadkilled Western Rattlesnakes
in broad age categories as detected in the White Lake Basin before (C) and after (D)
ecopassages were installed. Neonates are defined as rattlesnakes with snout-vent
lengths of < 30 cm, juveniles are 30-40 cm, subadults are 40-55 cm, and adults are >
55 cm…………………………………………………………………………………35
Figure 2.3. Figure 2.4. Jolly-Seber estimates (± 1 SE) for population size and survival of all
(top) and adult (bottom) Western Rattlesnakes (Crotalus oreganus) at focal dens in
the White Lake Basin, BC from 2015-2020. Also shown in the top graph is the
original Winton et al. (2020) population estimate for 2016 (■) as calculated using the
shorter mark/recapture history available during their study. Linear regression lines for
the population estimates of all snakes (y = 68.3x + 488.1, R2 = 0.33, P = 0.32) and
adults snakes (y = 4.9x + 309.3, R2 = 0.02, P = 0.81) are also shown. The solid arrow
represents the installation of snake-targeted ecopassages and the dashed arrow
indicates the installation of directional fencing as road mortality mitigation.….....…36

ix
Figure 3.1. Map of camera-monitored ecopassage locations on White Lake and Willowbrook
Roads, in the White Lake Basin, British Columbia, Canada. Ecopassages labelled in
red are existing drainage culverts, and ecopassages labelled in yellow were installed
in September 2017 at snake roadkill hotspots. Circles indicate the side of the road that
drift fencing was on. Lines representing ecopassages are not drawn to scale. (Base
map from Google Earth)…………………………......................................................48
Figure 3.2. Weekly ecopassage (n=8 ecopassages) camera appearances of Western
Rattlesnakes, Great Basin Gophersnakes, and Western Yellow-bellied Racers from
April 15 to October 6, 2019 and 2020 in the White Lake Basin of British Columbia,
Canada ……………………………………………………………………….............53
Figure 3.3. Proportion of appearances of Western Rattlesnakes, Great Basin Gophersnakes,
and Western Yellow-bellied Racers in 8 ecopassages in the White Lake Basin, British
Columbia, Canada in 2019 and 2020 combined. Dark-coloured bars represent newly
installed ecopassages, and light-coloured bars represent pre-existing culverts that
were modified with drift fencing. See Figure 3.1 for ecopassage locations…………54
Figure 3.4. Snake entrances through drift fenced versus unfenced ends of seven ecopassages
in 2019 and 2020 combined. The dotted line represents equal entrances from both
ends. Closed circles represent newly installed ecopassages, and open circles represent
pre-existing culverts that were modified with drift fencing.…....................................55

x
LIST OF TABLES
Table 3.1. Ecopassage camera appearances and passages of three ‘at-risk’ snake species in 8
ecopassages in the White Lake Basin, British Columbia, Canada. Percentages show
how many appearances resulted in passages (Passages/Appearances*100)…………51

Table 3.2. Ecopassage entrances of three ‘at-risk’ snake species detected by wildlife cameras
in 7 ecopassages in the White Lake Basin, British Columbia, Canada. One end of
each ecopassage is flanked by drift fencing, whereas the opposite entrance is not.
Paired T-tests were used to compare the number of fenced vs. non-fenced entrances
for each species in each of the two years of the study, and the resulting P values are
displayed here, with an asterisk (*) denoting a significant difference……………….52

Table A.1. Jolly-Seber population estimates for Western Rattlesnakes (Crotalus oreganus) in
the White Lake Basin, BC (2015-2020; estimates not available for 2015 or 2021
based on method of calculation)……………………………………………………..72

Table A.2. Jolly-Seber survivorship estimates for Western Rattlesnakes (Crotalus oreganus)
in the White Lake Basin, BC (2015–2021; estimates not available for 2020 or 2021
based on method of calculation)……………………………………………………..73

Table B.1. Frequency of roadkill detected for at-risk snake species during surveys and
incidentally along 11.7 kms of road within the White Lake Basin, BC,
Canada………………………………………………………………………………..74

Table B.2. Detections of dead Western Rattlesnakes (Crotalus oreganus) and densities per
km surveyed (unadjusted for observer error or scavenging rates) by road survey
method in the White Lake Basin, BC, Canada, 2015-2018………………………….75

Table B.3. Calculated road mortality rates (deaths/km/day) and number of Western
Rattlesnake (Crotalus oreganus) deaths based on walking survey results and
accounting for scavenger-removal and observer error during the active season (AprilOctober) along the 11.7 km survey route in the White Lake Basin, BC, Canada,
2015-2018……………………………………………………………………………76

xi
Table B.4. Adjusted dead Western Rattlesnake (Crotalus oreganus) road densities (dead
rattlesnakes/km) for five types of survey methods, accounting for scavenger-removal
and observer error, in the White Lake Basin, BC, Canada, 20152018…………………………………………………………………………………..77

Table C.1. The distribution of ecopassage use of Western Yellow-bellied Racers, Great Basin
Gophersnakes, and Western Rattlesnakes in the White Lake Basin, British Columbia,
Canada, with the proportion of the species total in brackets. An appearance was
defined as any camera detection of the species inside the ecopassage and a passage
occurred when a snake of the same species entered and exited opposite ends of the
ecopassage within 30 minutes………………………………………………………..78

Table D.1. Ecopassage camera appearances and passages of non-target species in 8
ecopassages in the White Lake Basin, British Columbia, Canada in 2019 and 2020.
(Adopted from Jaccard 2024).....…………………………………………………….79

CHAPTER 1
INTRODUCTION

Wildlife Mortality Factors
Although the mortality of individuals within animal populations may be attributable
to a single cause, often a number of natural and anthropogenic factors are working in tandem
with one another. Broadly put, these interactions may be additive or compensatory in nature.
Compensatory mortality is when mortality from one cause is reduced but the spared
individuals still end up dying from other causes (Mills 2013, Péron 2013), whereas additive
mortality is described as mortality compounded on to natural mortality rates, meaning
individuals would have otherwise survived if the proximate cause had been removed (Mills
2013, Péron 2013). Whether road mortality acts as a compensatory or additive force depends
on the population, but the latter is more likely for species with high adult survival rates and
delayed sexual maturity (Gibbs and Shriver 2002, Moore et al. 2023, Winton 2018).

Road Mortality and Mitigation
The presence of roads and vehicular traffic are one of the most wide-spread, global
threats to wildlife (Forman et al. 2003, Boyle et al. 2021). They are considered a main driver
of habitat degradation and destruction, posing a tremendous risk to wildlife both directly and
indirectly (Dillon et al. 2020, Trombulak and Frissell 2000). Direct mortality from vehicle
collisions can result in reductions in population persistence and viability (Boyle et al. 2021,
Fahrig and Rytwinski 2009).
In response to growing concern for wildlife conservation, mitigation measures are
being increasingly implemented to prevent road mortality (Glista e al. 2009, Jarvis et al.
2019, van der Grift et al. 2013). Some mitigation measures intended to reduce road-related
wildlife mortality are designed to influence motorist behaviour (e.g. warning signs, reduced
speed limits), while others seek to influence animal behaviour (e.g. fencing, crossing
structures), with the latter thought to be more effective (Huijser et al. 2007, Rytwinski et al.
12

13
2016). More broadly, wildlife crossing structures are intended not only to prevent direct
mortality by vehicles, but aid in maintaining population connectivity (Dillon et al. 2020).
They can take the form of under- or over-passes, and vary significantly in design depending
on the intended species, with some examples being amphibian tunnels, badger pipes, wildlife
bridges, rope bridges, and glider poles (Glista et al. 2009, Rytwinski et al. 2016, van der Grift
et al. 2013). ‘Ecopassage’ is a term used for under-road tunnels that often are used to
facilitate road-crossing movements in smaller animals, and they are becoming an
increasingly common mitigation tool for the conservation of reptile species (Baxter-Gilbert et
al. 2015, Boyle et al. 2021, Dillon et al. 2020).
Despite ecopassages now being used more commonly for reptile conservation, indepth analyses of their effectiveness for this taxon remain rare (Baxter-Gilbert et al. 2015,
Boyle et al. 2021, Dillon et al. 2020). The research to date most often assesses only the
amount of roadkill or the amount the ecopassage is used, but these metrics do not necessarily
equate to effectiveness (van der Grift et al. 2013). Changes in individual behaviour can affect
survival and reproduction, which in turn ultimately affects populations (French et al. 2018).
For a more robust and accurate assessment of the effectiveness of these mitigation structures
for reptiles, roadkill, ecopassage use, and population-level impacts should be considered
(Boyle et al. 2021, Fahrig and Rytwinski 2009).

Snakes and Roads
Reptiles are among the fastest-declining taxa globally, facing numerous wide-spread
threats including habitat loss, degradation, and fragmentation (Böhm et al. 2013, Cox et al.
2022, Saha et al. 2018). Roads are an example of these threats, and reptiles are particularly
vulnerable to their effects due to their small size and slow rates of movement and
reproduction (Boyle et al. 2021, Brehme et al. 2018, Gigeroff and Blouin-Demers 2023).
Within the reptiles, snakes are particularly at-risk of direct road mortality for a
number of reasons. Seasonal migrations to and from hibernacula put snakes at risk of
encountering roads, especially if hibernacula lie in close proximity to a road (Gunson and
Schueler 2019). Snakes also are drawn to roads to thermoregulate (Baxter-Gilbert et al. 2015,
Mccardle and Fontenot 2016) and morphological factors, like an elongated body shape,

14
creates roadkill risk for snakes, such as when motorists may fail to detect small or juvenile
animals or mistake them for sticks or twigs. (Gunson and Schueler 2019). Unfortunately,
snakes also are intentionally run over by motorists (Ashley et al. 2007). Species with life
histories characterized by low reproductive rates and low adult mortality, like rattlesnakes,
are particularly vulnerable to demographic consequences of road mortality (Brehme et al.
2018, Forman et al. 2003).
These effects are amplified in northern snake species. In British Columbia for
example, rattlesnakes face shorter cooler active seasons, that limit their annual growth rate
and reproductive frequency even further (Macartney and Gregory 1988). They also are
restricted to the southern latitudes of the political borders of Canada while being at the
northern limits of their geographic range more than any other taxa (Currie and Marconi
2020). Species that are not limited by density dependence, and instead limited by intrinsic
factors like growth rate and reproductive frequency, will not experience a strong
compensatory response to anthropogenic mortality since the presence of other individuals is
not restricting their ability to survive (Péron 2013). Therefore, road mortality generally is an
additive factor that can be detrimental to northern snake populations (Winton et al. 2020).

Study Site
The Okanagan is home to the fastest-growing metropolitan area in Canada (Statistics
Canada 2022), and is a tourism hotspot (Destination British Columbia 2017). As a result,
there has been immense expansion of infrastructure and therefore, loss or degradation of
wildlife habitat. This is concerning since the South Okanagan is where the northern edge of
the Great Basin Desert reaches into Canada, making it an extremely rare environment that is
home to many species occurring nowhere else in the country (B.C. Ministry of Environment
1998, Parks Canada 2022).
Within the South Okanagan river valley is the White Lake Basin, which includes part
of the White Lake Grasslands Protected Area. This landscape is home to many species at risk
(BC Parks 2023, Figure 1.1), including three federally Threatened snake species: Western
Rattlesnakes (Crotalus oreganus), Great Basin Gophersnakes (Pituophis catenifer

15
deserticola), and Western Yellow-bellied Racers (Coluber constrictor mormon). Historical
work by Winton et al. (2018, 2020) focused on Western Rattlesnake road mortality and its
impacts on the population in the area. This area and the species make an ideal combination
for such a study, since there are very few anthropogenic features on the landscape apart from
roads, and the communal denning behaviours of rattlesnakes make an extensive markrecapture study achievable (Macartney et al. 1990). More specifically, the Winton research
accurately quantified rattlesnake road mortality by testing and correcting for associated
sources of error, characterized the rattlesnake population affected by road mortality, and
assessed the long-term persistence of the population under the threat of roadkill. They found
~6.6% of the rattlesnake population dying each year from this road mortality, and using a
Population Viability Analysis they determined that even a slight increase in road mortality
could result in the species becoming extirpated from the area within 100 years (Winton et al.
2020). They recommended establishing mitigation measures in the area as an effort to
conserve the species, which resulted in the installation of ecopassages and drift fencing.

Research Objectives
In this thesis I assessed the short-term impact of ecopassages on a Western
Rattlesnake (Crotalus oreganus) population in British Columbia, Canada. I conducted
intensive surveys to monitor roadkill and assess changes in population size. By repeating the
same methodology used by Winton et al. (2018, 2020) to estimate roadkill occurrences,
population size, and annual survivorship, I was able to compare these metrics for the two
years immediately following the establishment of the ecopassages and associated drift
fencing to the three years before in which Winton et al. conducted the same measurements.
I also quantified and compared the use of eight ecopassages in the area by Western
Rattlesnakes (Crotalus oreganus), Great Basin Gophersnakes (Pituophis catenifer
deserticola), and Western Yellow-bellied Racers (Coluber constrictor mormon). I positioned
wildlife cameras at either end of each ecopassage to monitor the occurrence of each species
throughout their active season, enabling me to compare use both spatially and temporally.
The overarching research objectives of my thesis were to:

16
1. Compare Western Rattlesnake pre-mitigation roadkill and population estimates to
those in the years immediately following the installation of ecopassages and fencing;
and
2. Quantify ecopassage use of three at-risk snake species, considering spatial and
temporal differences between species.
In the remaining portion of this chapter, I provide an overview of my study site and more
information about the ecopassages and drift fencing that were installed there. I collected data
for this research in 2019 and 2020, but data from all years (since 2015) of this long-term
project were incorporated into this thesis. In Chapter 2, I replicate the work done by Winton
et al. (2018, 2020) to assess the immediate impact of the newly installed mitigation structures
on the road mortality rates and population size of rattlesnakes in the area. In Chapter 3, I use
photos from wildlife cameras to quantify and compare the ecopassage use of Western
Rattlesnakes, Great Basin Gophersnakes, and Western Yellow-bellied Racers. I also conduct
a secondary assessment of fence effectiveness. Finally, in Chapter 4, I summarize the
findings of my research and discuss the resulting management implications. I conclude with
highlighting the importance of both short- and long-term data sets in fully understanding the
response of these animals to mitigation efforts.
Site Description
My study took place in the White Lake Basin of the South Okanagan region of British
Columbia, Canada (latitude 49.318N, longitude 119.638W). The basin consists of open
shrub-steppe grassland habitat, and is managed with the objective to “integrate livestock
management with conservation of habitat for species at risk” by The Nature Trust of British
Columbia (TNTBC 2022). The paved, undivided two-lane road (BC Class 5 highway)
traversing the bottom of the basin bisects the grassland and has an unposted speed limit of 80
km/h, in part due to the presence of a Canadian federal astrophysical observatory. Two roads
– one running largely East-West (White Lake Road) and one running North-South
(Willowbrook Road) – meet at a T-intersection (Figure 1.1). There is barbed-wire fence

Figure 1.1. Study site location within the Western Rattlesnake (Crotalus oreganus) range in
western North America (from Winton 2018).
17

running ≈ 5m parallel to the road, and each year the vegetation along the road shoulder is
mowed to a height of less than 0.5 m for a distance of 1.8 m from the road edge. The basin
largely lacks infrastructure apart from the roads and the Dominion Radio Astrophysical
Observatory (run by the National Research Council Canada). The basin is surrounded by
rolling hills, steep bluffs, and two golf courses with small residential communities and ponds
lie just outside the basin. Elevation within the basin ranges from 500 – 1100 m.
The South Okanagan valley is characterized by hot, dry summers and long, cold
winters with temperatures reaching below freezing (Meidinger and Pojar 1991). The average
temperature patterns were consistent throughout the years of the study, although February
2019 was colder than usual. The total monthly precipitation patterns also were relatively
similar with more variation, particularly in 2017, which had high amounts of precipitation in
May and none in July or August (Figure 1.2).
Initial research on Western Rattlesnake road mortality in the White Lake Basin began
in 2015, focusing on road surveys and extensive mark-recapture efforts (Winton et al. 2018,
2020, Figure 1.3). New ecopassages were installed in 2017, and cameras were mounted
inside the entrances in 2018. The first two years to have cameras deployed in the ecopassages
throughout the entire active season were 2019 and 2020. Drift fencing was installed at
ecopassage entrances in early 2019 (Figure 1.4). Details of these structures can be found in
the Methods section of Chapter 3.
The second year of this study took place in the summer of 2020 during the height of
the Covid-19 pandemic, which admittedly may have altered traffic patterns in the White Lake
Basin (Destination British Columbia 2017). However, field methods and the collection of
traffic data were conducted consistently with other years of the study.

18

19

30

A

2015
2016
2017
2018
2019
2020
2021

25

Temperature (°C)

20
15
10
5
0
-5
-10
Jan

Feb

Mar

Apr

May

Jun

Jul

Aug

Sep

Oct

Nov

Dec

Month

100

B

2015
2016
2017
2018
2019
2020
2021

Precipitation (mm)

80

60

40

20

0

Jan

Feb

Mar

Apr

May

Jun

Jul

Aug

Sep

Oct

Nov

Dec

Month

Figure 1.2. A) Mean monthly temperature (°C) and B) total monthly precipitation (mm) in
the South Okanagan during the study years (2015-2021) as measured at the Penticton
Regional Airport, British Columbia, Canada (49°N, 119°W; Environment and Climate
Change Canada 2024).

Figure 1.3. Timeline of snake research in the White Lake Basin, British Columbia.

20

A

B

C

D

Figure 1.4. Examples of snake road mortality mitigation measures in the White Lake Basin, British Columbia, including A) a modified
culvert equipped with a camera, with the entrance unaccompanied by fencing, B) a newly-installed semi-circular ecopassage equipped
with a camera and fencing surrounding the entrance, and C & D) drift fencing positioned to funnel snakes into under-road
ecopassages. (Photo credits: Karl Larsen)
21

22

LITERATURE CITED
Aresco MJ. 2005. Mitigation measures to reduce highway mortality of turtles and other
herpetofauna at a north Florida lake. Journal of Wildlife Management. 69(2): 549560.
Ashley EP, Kosloski A, Petrie SA. 2007. Incidence of intentional vehicle-reptile collisions.
Human Dimensions of Wildlife. 12(3): 137-143.
Baxter-Gilbert JH, Riley JL, Lesbarrères D, Litzgus JD. 2015. Mitigating reptile road
mortality: fence failures compromise ecopassage effectiveness. PLoS ONE 10(3).
B.C. Ministry of Environment. 1998. What makes the Okanagan a special area? in Habitat
Atlas for Wildlife at Risk. Government of British Columbia; [accessed 2024 Feb 14].
Available from: https://www.env.gov.bc.ca/okanagan/esd/atlas/special_area.html
BC Parks. c2023. White Lake Grasslands Protected Area. Province of British Columbia;
[accessed 2024 Feb14]. Available from: https://bcparks.ca/white-lake-grasslandsprotected-area/
Böhm M, Collen B, Baillie JEM, Bowles P, Chanson J, Cox N, Hammerson G, Hoffmann M,
Livingstone SR, Ram M, et al. 2013. The conservation status of the world’s reptiles.
Biological Conservation. 157: 372-385.
Boyle SP, Keevil MG, Litzgus JD, Tyerman D, Lasbarréres D. 2021. Road-effect mitigation
promotes connectivity and reduces mortality at the population-level. Biological
Conservation. 261.
Brehme CS, Hathaway SA, Fisher RN. 2018. An objective road risk assessment method for
multiple species: Ranking 166 reptiles and amphibians in California. Landscape
Ecology. 33: 911-935.
Cox N, Young BE, Bowles P, Fernandez M, Marin J, Rapacciuolo G, Böhm M, Brooks TM,
Hedges SB, Hilton-Taylor C, et al. 2022. A global reptile assessment highlights
shared conservation needs of tetrapods. Nature. 605(7909): 285-290.
Currie J, Marconi V. 2020. An analysis of threats and factors that predict trends in Canadian
vertebrates designated as at-risk. FACETS 5: 49–66.
Destination British Columbia. 2017. Thompson Okanagan Regional Tourism Profile.
Vancouver, BC; [accessed 2023 Oct 11]. Available from:
https://www.destinationbc.ca/content/uploads/2018/05/Thompson-OkanaganRegional-Tourism-Profile_2017.pdf
Dillon RM, Boyle SP, Litzgus JD, Lesbarrères D. 2020. Build it and some will use it: A test
of road ecopassages for Eastern Gartersnakes. Journal of Herpetology. 54(1): 19–23.

23
Environment and Climate Change Canada. 2024. Historical Climate Data. [accessed 2024
Feb 15]. Available from: http://climate.weather.gc.ca/
Fahrig L, Rytwinski T. 2009. Effects of roads on animal abundance: An empirical review and
synthesis. Ecology and Society. 14(1): 21.
Forman RT, Sperling TD, Bissonette JA, Clevenger AP, Cutshall CD, Dale VH, Fahrig L,
France R, Goldman CR, Heanue K, et al. 2003. Road Ecology: Science and Solutions.
Washington (DC): Island Press. 481 p.
French SS, Webb AC, Hudson SB, Virgin EE. 2018. Town and country reptiles: A review of
reptilian responses to urbanization. Integrative and Comparative Biology. 58(5): 948966.
Gibbs JP, Shriver WG. 2002. Estimating the effects of road mortality on turtle populations.
Conservation Biology. 16(6): 1647-1652.
Gigeroff AES, Blouin-Demers G. 2023. Do roads affect the abundance of garter
(Thamnophis sirtalis) and redbelly snakes (Storeria occipitomaculata)? Canadian
Journal of Zoology. 101: 267-275.
Glista DJ, DeVault TL, DeWoody JA. 2009. A review of mitigation measures for reducing
wildlife mortality on roadways. Landscape and Urban Planning. 91: 1-7.
Gunson KE, Schueler FW. 2019. Wildlife on Roads: A Handbook. Peterborough (ON): EcoKare International. 232 p.
Huijser MP, McGowen PT, Fuller J, Hardy A, Kociolek A, Clevenger AP, et al. 2007.
Wildlife-vehicle collision reduction study. Report to U.S. Congress. U.S. Department
of Transportation, Federal Highway Administration, Washington DC, USA.
Jarvis LE, Hartup M, Petrovan SO. 2019. Road mitigation using tunnels and fences promotes
site connectivity and population expansion for a protected amphibian. European
Journal of Wildlife Research. 65: 27.
Macartney JM, Gregory PT. 1988. Reproductive biology of female rattlesnakes (Crotalus
viridis) in British Columbia. Copeia. 1988(1): 47–57.
Macartney JM, Gregory PT, Charland MB. 1990. Growth and sexual maturity of the Western
Rattlesnake, Crotalus viridis, in British Columbia. Copeia. 1990(2): 528-542.
Markle CE, Gillingwater SD, Levick R, Chow-Fraser P. 2017. The true cost of partial
fencing: Evaluating strategies to reduce reptile road mortality. Wildlife Society
Bulletin. 41(2): 342-350.
Mccardle LD, Fontenot CL. 2016. The influence of thermal biology on road mortality risk in
snakes. Journal of Thermal Biology. 56: 39-49.

24
Mills LS. 2013. Conservation of Wildlife Populations: Demography, Genetics, and
Management. 2nd edition. Chichester (UK): John Wiley & Sons, Ltd. 354 p.
Moore LJ, Petrovan SO, Bates AJ, Hicks HL, Baker PJ, Perkins SE, Yarnell RW. 2023.
Demographic effects of road mortality on mammalian populations: A systematic
review. Biological Reviews. 98: 1033-1050.
Parks Canada. 2022. Proposed National Park Reserve in the South Okanagan-Similkameen.
Government of Canada; [accessed 2024 Feb 14]. Available from:
https://parks.canada.ca/pn-np/cnpn-cnnp/okanagan
Péron G. 2013. Compensation and additivity of anthropogenic mortality: Life-history effects
and review of methods. Journal of Animal Ecology. 82: 408-417.
Read KD, Thompson B. 2021. Retrofit ecopassages effectively reduce freshwater turtle road
mortality in the Lake Simcoe Watershed. Conservation Science and Practice. 3(2).
Rytwinski T, Soanes K, Jaeger JAG, Fahrig L, Findlay CS, Houlahan J, van der Ree R, van
der Grift EA. 2016. How effective is road mitigation at reducing road-kill? A metaanalysis. PLoS ONE. 11(11).
Saha A, McRae L, Dodd Jr. CK, Gadsden H, Hare KM, Lukoschek V, Böhm M. 2018.
Tracking global population trends: Population time-series data and a living planet
index for reptiles. Journal of Herpetology. 52(3):259-268.
Statistics Canada. 2022. Table 98-10-0005-01. Population and dwelling counts: Canada,
provinces and territories, census metropolitan areas and census agglomerations. DOI:
https://doi.org/10.25318/9810000501-eng
Trombulak SC, Frissell CA. 2000. Review of ecological effects of roads on terrestrial and
aquatic communities. Conservation Biology. 14(1): 18-30.
van der Grift EA, van der Ree R, Fahrig L, Findlay S, Houlahan J, Jaeger JAG, Klar N,
Madriñan, Olson L. 2013. Evaluating the effectiveness of road mitigation measures.
Biodiversity Conservation. 22: 425-448.
Winton SA. 2018. Impacts of road mortality on the Western Rattlesnake (Crotalus oreganus)
in British Columbia [MSc thesis]. Kamloops, BC, Canada: Thompson Rivers
University.
Winton SA, Bishop CA, Larsen KW. 2020. When protected areas are not enough: Lowtraffic roads projected to cause a decline in a northern viper population. Endangered
Species Research. 41:131–139.
Winton SA, Taylor R, Bishop CA, Larsen KW. 2018. Estimating actual versus detected road
mortality rates for a northern viper. Global Ecology and Conservation. 16 (2018).

25

CHAPTER 2
TRANSITIONAL RESPONSE OF WESTERN RATTLESNAKES (CROTALUS
OREGANUS) TO THE INSTALLATION OF ROAD ECOPASSAGES AND

DRIFT FENCING
INTRODUCTION
Roads and traffic are a major threat to wildlife populations across many taxa, both
through direct mortality and the fragmentation of habitat, the latter creating barriers within
populations (Baxter-Gilbert et al. 2015, Fahrig and Rytwinski 2009). Reptiles are one of the
fastest-declining taxa globally, being particularly vulnerable to road effects due to their small
size, attraction to roads for thermoregulation, and low vehicle avoidance (Baxter-Gilbert et
al. 2015, Fahrig and Rytwinski 2009). Reptile species with life histories characterized by low
reproductive rates and low adult mortality are even more vulnerable to the demographic
consequences of road mortality (Boyle et al. 2021, Forman et al. 2003). These impacts can be
exacerbated in reptile populations at higher latitudes that experience additional constraints
due to climate (Macartney et al. 1990).
To reduce road mortality while improving habitat connectivity for vulnerable reptile
species, the implementation of mitigation measures has become increasingly common (Jarvis
et al. 2019, Dillon et al. 2020). Ecopassages are commonly installed under roads to facilitate
the crossing of roads, and often are coupled with drift fencing meant to direct wildlife into
the passageways. Such structures usually are targeted towards recovering a specific species at
risk. However, despite the increasing application of these mitigative tactics for reptile
survival, in-depth analyses of their effectiveness remain rare, both in terms of their direct
impact on roadkill rates as well as the persistence of populations or communities (Boyle et al.
2021, Fahrig and Rytwinski 2009).
Intuitively, the response of wildlife to newly-installed mitigation structures will vary
in time following the moment of installation. Whelan et al. (2002) outlined how population
changes following wildfire will depend on the organism, and their species-specific mortality,
reproductive rates, immigration, and emigration. These responses also are likely to be age- or

26
stage- specific, being highly linked with the life cycles of the organism as well as the timing
and severity of the disturbance. In referring to the effects of wildfire, Warren et al. (1987)
termed the period immediately following disturbance the ‘shock phase’ of ecosystem
recovery, followed by the recovery phase. Analogous effects may be associated with the
implementation of anthropogenic structures designed to reduce road impacts on wildlife:
their sudden presence on the landscape likely will impact species differently, leading to
‘stages’ in the recovery of populations and communities. For example, crossing structure use
increases over time for large mammals, as animals become acclimatized and learn to start
using them (Ford et al. 2017, Gilhooly et al. 2019, Seidler et al. 2018).
Similar research on the response of reptiles and other small-bodied animals to the
addition of road crossing structures is lacking (Baxter-Gilbert et al. 2015, Boyle et al. 2021,
Dillon et al. 2020). Initiating monitoring programs immediately after mitigation is affected
allows the ‘shock phase’ response to be examined, while providing a foundation for longerterm monitoring. Knowledge of if and when these structures become effective at reducing
roadkill, or how responses will vary through time, becomes essential for designing
monitoring programs and/or interpreting results at different stages of response.
I postulate that snakes are a taxa where initial responses to new structures, such as
mitigative ecopassages, may pass through a ‘shock phase’. Conspecific scent trailing is
important for snakes, in particular communally aggregating snakes (Brown and MacLean
1983, Muellman et al. 2018). Therefore, it may take time for snakes to initially find these
new corridors and establish repetitive and frequent use of structures and pathways. Snakes
demonstrating large, annual migration distances, fidelity to their overwintering and summer
habitat, and a tendency to take the same path and re-visit specific habitat features from year
to year likely are especially reliant on scent trails (Duval et al. 1990, Gomez et al. 2015,
Parker and Anderson 2007). These conditions likely will result in an even more pronounced
‘shock phase’.
As rattlesnakes exhibit the characteristics mentioned above, they may demonstrate a
muted response to mitigation structures during the ‘shock phase’. Establishing scent trails in
the ecopassages likely is important for other rattlesnakes to begin using them since neonates
of some rattlesnake species follow scent trails of conspecifics (Brown and MacLean 1983,
Muellman et al. 2018, Scudder et al. 1988). Furthermore, since vipers take relatively longer

27
to transform energy from prey into offspring than similar-sized endotherms (i.e. longer
generation times), the result is a slower population-level response to changes in their
environment (Nowak et al. 2008). This slow response time likely is amplified in northern
viper populations that face shorter and cooler active seasons, such as rattlesnakes living in
Canada (Macartney and Gregory 1988).
Winton et al. (2018, 2020) collected baseline data on the impacts of traffic on a rural
community of Western Rattlesnakes (Crotalus oreganus) in British Columbia, including
demographic changes. Their work documented significant road mortality in the population (≈
7% annually) despite low traffic (350 vehicles/day) on roads bisecting an otherwise
reasonably-intact ecosystem. The Winton study also identified specific roadkill ‘hotspots’
along the two roads in the area, where relatively high mortality appeared linked to migratory
movement corridors to and from hibernacula (Winton 2017). In an effort to curb road
mortality, eight ecopassages were established within the study area at these locations (see
Figure 3.1).
Using identical survey methods to Winton et al. I assessed the immediate impact
(during what is likely a ‘shock phase’) of the ecopassages on the road mortality rates of
rattlesnakes in the same study area. By repeating the same walking surveys and methodology
used by Winton et al. (2018, 2020), I examine changes in road mortality and population
demographics during the years immediately following the establishment of the ecopassages
and associated drift fencing. I predicted I would not detect any significant changes in roadkill
rates or population size during this two year time period post-implementation of ecopassages
and drift fencing, due to the limited amount of time available for the snakes to adjust to the
structures.

METHODS
Study Site
This study was conducted in the White Lake Basin (latitude 49.318N, longitude
119.638W) in the South Okanagan region of British Columbia, Canada (Winton et al. 2018,
2020). Here, near the northern limit of the Western Rattlesnake’s range (see Figure 1.1) the

28
typical annual activity period of the snake generally spans April – October. The basin
consists of open shrub-steppe grassland habitat, surrounded by rolling hills and steep bluffs.
The area is characterized by Big Sagebrush (Artemisia tridentata), Bluebunch Wheatgrass
(Agropyron spicatum), and Ponderosa Pine (Pinus ponderosa) (Meidinger and Pojar 1991).
The area is managed with the objective to “integrate livestock management with conservation
of habitat for species at risk” by The Nature Trust of British Columbia (TNTBC 2022).
Traversing the valley bottom is a paved, undivided two-lane road (BC Class 5 highway) with
an unposted speed limit of 80 km/h.
In September 2017, four ecopassages were newly installed through the study area at
identified snake roadkill hotspots (Winton 2017), and modifications were made to an
additional four existing drainage culverts. This produced eight ecopassages within a 6.5 km
stretch of road, with the intended purpose of lowering Western Rattlesnake road mortality.
All eight ecopassages were made from corrugated metal, although, the newly-installed
ecopassages were more oval-shaped than the older, rounder drainage culverts (See Figure
1.4. A & B). The ecopassages had an average length of 12 ± 1 SD m, height of 45 ± 8 SD
cm, and width of 67 ± 12 SD cm, with an average openness (= height * width / length) of 2.5
± 0.5 SD cm. A substrate of sand (≈ 5 cm deep) lined the bottom of the culverts.
In April 2019, drift fencing was added to some of the entrances of the ecopassages,
with a design specifically targeted towards directing snake movement (model AMX-SP40,
Animex Wildlife Fencing Solutions, see Figure 2.1) into the ecopassages (Baxter-Gilbert et
al. 2015, Colley et al. 2017). Construction of the fences was hampered by bedrock, drainage
patterns, and cattle use, thus preventing a robust experimental design, such as altering
fencing on upslope versus downslope entrances of the ecopassages. Still, drift fences
(varying lengths, and ≈ 75 cm in height) were established at one end of each of seven
ecopassages (See Figure 1.4. C & D, and Figure 3.1), while one received no fencing. Fence
lengths flanking the entrances of the ecopassages ranged from a total length of 43-63 m, with
an average length of 51.5 ± 8.3 SD m. A total of 0.4 km of the 23.4 km of road edge within
the study area were fenced.

29
Calculating Road Mortality Rates
To determine roadkill rates, I duplicated the survey method established by Winton et
al. (2018). An 11.7 km route along White Lake Road and Willowbrook Road was surveyed
approximately every 3 days during the active season, from April to October. Road surveys
were conducted on foot: two observers walked along opposite edges of the road, scanning for
dead and alive snakes on the road and 1.8 m vegetation control zone on the road shoulder.
Throughout the duration of the research at this site road survey effort (total kilometers
surveyed) varied from year to year, with a minimum of 465 kms (2015), a maximum of 865.8
kms (2017), and an average of 625.7 kms surveyed.
Also following Winton et al. (2018), I conducted four observer-bias experiments to
calculate the detection probabilities of dead snakes during road surveys. A third researcher
planted dead snakes (previously roadkilled) of a variety of sizes on the road and shoulder at
sites along the road survey route, and revisited these sites immediately following the survey
(within approximately 30 minutes) to determine whether the observers (hereon known as
surveyers) had detected or missed each carcass. Four observer detection trials were
completed during walking road surveys from July-September, 2020. Detection probability
was calculated using the proportion of planted snakes observed by the surveyers, which then
were used in the model developed by Winton et al. (2018) to provide a more accurate
roadkill estimate.
Detections of dead rattlesnakes on the road survey route outside of official survey
times also were recorded, and combined with the gross number of road survey detections to
provide a number of total detections for that year. Two traffic counters (TRAFx, G4) – one
on each of the two roads within the survey route – were provided by the BC Ministry of
Transportation and Infrastructure (MoTI) to provide daily traffic volumes. Willowbrook
Road was consistently busier than White Lake Road, so data from the traffic counter on
Willowbrook Road was used to calculate the mean number of vehicles per day in the study
site (maximum daily average) for the time period June 1 – August 31 of each year, when
traffic volumes typically peak in the region due to school closures and tourist travel.
Mortality rates due to roadkill were modelled following Winton et al. (2018): the
density of dead rattlesnakes on the road, observer detection error, rate of carcass removal by

30
scavengers, and time since the last survey were combined to calculate a rate with units of
deaths/km/day. I conducted a preliminary analysis of the trend in snake mortality rates using
linear regression.

Monitoring Population Trajectory
To track the rattlesnake population in the basin, as per Winton et al. (2020), I
conducted mark-recapture work at 6 focal hibernacula located within 400 m of the road.
Repeated visits to these hibernacula occurred during spring egress and fall ingress from
Spring 2015 to Autumn 2021. Marking of snakes was done using implantable PIT (passive
integrated transponder) tags (Mini HPT8 PIT Tag, Biomark, Inc.). The general area also was
arbitrarily surveyed throughout the active season for snakes from these dens.
Appending my extended mark-recapture data with that collected by Winton et al.
(2020), I produced updated Jolly-Seber population and survival estimates of the rattlesnake
population using data from the 2015-2021 capture sessions. Jolly-Seber population estimates
cannot be calculated for the first and final sampling periods in a data set, and additionally,
survival estimates cannot be calculated for the second last and final periods (Krebs 1989):
therefore, the total of seven capture sessions provided five Jolly-Seber population estimates
(2016-2020) and 4 survival estimates (2015-16 to 2018-19). Estimates were generated using
the Rcapture package (Baillargeon and Rivest 2007) for RStudio (version 4.1.1; R Core
Team, 2021), with each active season (April-October) being considered a ‘capture session’. I
used linear regression to conduct preliminary analyses of trends in the yearly population
estimates.

RESULTS
Road Surveys
A total of 53 road surveys were conducted in 2019 and 49 in 2020, with an average of
2.6 and 3.2 days between surveys, respectively. Between road surveys and incidental

31
observations outside of the surveys, a total of 35 dead rattlesnakes were found within the
survey route in 2019, and 21 in 2020. Analogous detections from research done in 2015-2018
were 36, 56, 28, and 29, in each year respectively (2015-2016 from Winton et al. 2018, 20172018 from Larsen, unpublished data).
Traffic Rates
Traffic rates shown in Figure 2.1 represent the daily average number of vehicles on
Willowbrook Road (the busier of the two roads within the survey route) from June-August.
2015 was the year with the lowest traffic rate of 266 vehicles per day, and 2020 was the
highest with a rate of 505 vehicles per day. Unfortunately, data for 2018 were lost when the
traffic counters were removed prematurely by government personnel.

Roadkill Rates
Roadkill rates show a trend suggestive of a decline over the study years, despite
traffic rates increasing (Figure 2.1). The linear regression results indicate that this trend is
insignificant, but this analysis should be considered preliminary due to the insufficient
sample size of six years. The year 2018 had low roadkill rates but unfortunately the missing
traffic data makes it impossible to comment on how the rates in that year were related to
traffic.

Mark-Recapture
Total snakes captured (new individuals + recaptures) ranged from 172 to 324
individuals per year across all study years. In the pre-mitigation years of the study
(2015/2016) mid-sized adults in the 60-70cm SVL size category were the greatest proportion
of captures, with neonates in the 20-30 cm category having about half as many captures. In
the post-mitigation years of the study (2019/2020) the proportion of neonate captures was
higher than any other individual size category, resulting in the size distribution of snake
captures being inconsistent in the pre- and post-mitigation years (Figure 2.2. A & B, Twosample Kolmogorov-Smirnov test, D = 0.191, P < 0.001). Also, there were many fewer

32
roadkilled neonate carcasses detected post-mitigation despite the high proportion of neonate
captures in that time period (Figure 2.2 B & D). Roadkilled neonates were classified more
broadly in age categories (versus SVL size categories), as carcasses are often so damaged
that they are much more difficult to accurately measure than live snakes. For these reasons, I
limit the comparison over time to only adult snakes, where there was less variation in the
proportion of snakes captured from year to year.

Population Estimate
Jolly-Seber population estimates for all snakes and adult snakes were the highest in
2019 (900 ± 96 SE and 405 ± 47 SE, respectively) and lowest in 2017 (526 ± 54 and 272 ±
25). Population and survival estimates for all snakes appear in Figure 2.3 (top), alongside the
original population estimate from Winton et al. (2020). Figure 2.3 (bottom) displays the
Jolly-Seber population and survival estimates for adult rattlesnakes only (≥55 cms). In both
cases population estimates were higher during the period when mitigation structures are
being installed, but then lower in 2020. The linear regression results indicate that the
population trends for all snakes and adult snakes are insignificant, but these results should be
considered preliminary due to the insufficient sample size of five years.

DISCUSSION

This study highlights the necessity of monitoring mitigation effects both before and
after implementation, given the difficulty of drawing firm conclusions based on short-term
data. Although there was a suggestion of decline in rattlesnake roadkill rates despite an
increase in traffic rates post-mitigation, it may be too early to detect concrete shifts in
population size and survival rates. Still, this study provides an important snapshot of the
‘shock phase’ response to mitigation structures, and a strong foundation for monitoring
beyond it (Wong in prep).

33
The two post-mitigation years have some of the lowest road mortality rates, but the
highest traffic rates. Camera data (see Chapter 3) confirms that rattlesnakes were using the
ecopassages in some capacity, and therefore these results could be due to ecopassage use by
the snakes. However, road avoidance by rattlesnakes is another possible explanation. Other
taxa have demonstrated increased road avoidance with increased traffic volume (Jacobson et
al. 2016, Loraamm et al. 2021). Western Rattlesnakes in British Columbia display smaller
home ranges and shorter range lengths in disturbed areas compared to those in areas with no
human disturbance (Lomas et al. 2019). Another Canadian rattlesnake species (Sistrurus
catenatus catenatus) moves shorter distances and less frequently with increased human
disturbance, defined by a higher number of vehicles on roads and more hikers (Parent and
Weatherhead 2000). If higher volumes of traffic are causing limited movement of Western
Rattlesnakes around the road, it could lead to barrier effects that can have significant impacts
on populations. These include reduced gene flow, fragmentation of populations close to the
road, and altered habitat use that increases energetic costs (Forman et al. 2003, Paterson et al.
2019). Since northern snake populations already face environmental factors that limit growth
and reproductive frequency, the added energetic costs of road avoidance could be detrimental
to their persistence (Macartney and Gregory 1988). Western Rattlesnake populations in BC
have significant genetic differentiation and limited population connectivity in areas with
major highways (Schmidt et al. 2019), so future development around the basin could further
increase traffic volumes, exacerbating the situation.
The positive Jolly-Seber population and survivorship estimate trends represent a relatively
short period, but are still encouraging. Although these data suggest an increase in rattlesnake
population size after the ecopassage installation in 2017, it is unlikely that a population-level
change of that magnitude in such a short time frame is directly attributable to the mitigation
measures, particularly when a population drop occurs again in 2020. Snake populations can
naturally fluctuate over time due to both density dependence and weather-induced factors
affecting juvenile survival or prey availability (Altwegg et al. 2005, Madsen et al. 2006,
Shine et al. 2021), but these factors were not measured in this study. Northern vipers exhibit
more “K-selected” life-history traits than other snakes, including delayed maturation,
infrequent reproduction, and low fecundity (Macartney and Gregory 1988). This results in

34
more density-dependent population growth that typically would not express itself as rapid
and drastic fluctuations (Shine et al. 2021). Therefore, environmental factors also are likely
to be contributing to these changes in population. Although our data set is relatively
extensive for this type of assessment, the variation in these population parameters suggests
long-term monitoring will be necessary to conclude if there are population-level impacts of
these mitigation measures. Currently, a generation for Western Rattlesnakes in British
Columbia is estimated at approximately 13.7 years (Maida et al. 2018). Therefore, one would
expect a detectable response in the population to take longer to manifest than the time period
encompassed in this study. Increasing use of crossing structures over time has been
documented in mammals (Ford et al. 2017, Gilhooly et al. 2019, Seidler et al. 2018), but
remains to be clearly shown in reptiles.
The population estimates calculated using only adult rattlesnakes are a better
representation of the trend throughout the duration of the study than those using all the
snakes. The higher proportion of neonates captured in the post-mitigation years (2019 and
2020) is possibly due to a difference in sampling effort or marking method from the premitigation years. Initially scute clipping was used to identify neonate rattlesnakes because
PIT tag needles were too large to be used on them. Scute clips are visible for a short time, but
as the snake grows and sheds they become difficult to distinguish (Fitch 1987). As smaller
PIT tags became available in the later years of the study, we began PIT tagging all neonate
rattlesnakes, which is a more reliable form of identification (Jeminson et al. 1995).
Therefore, a scute clipped neonate caught later as an adult may have been considered a new
individual, contributing to inaccuracies in the ‘all snakes’ survivorship and population
estimates. Regardless, since neonate survival appears low (6 – 46%; Macartney 1985), using
only adults for this type of analysis likely is a more accurate predictor of functioning
population size. The capture and marking of relatively larger numbers of neonates, such as
seen in 2019 and 2020, would inflate the population estimate, given age class survival rate is
not factored into a Jolly-Seber population estimate.
Should a more pronounced decline in roadkill and an increase in the population size
be expected as monitoring continues beyond the ‘shock phase’? In other species, snakes

Figure 2.1. Calculated Western Rattlesnake (Crotalus oreganus) roadkill rates (± 1 SE) (left ordinate, black data points) with a linear
regression line (y = -0.006x + 0.0613, R2 = 0.42, P = 0.17), and average traffic rates (right ordinate, bar graph) from 2015-2020.
Traffic data were unavailable for 2018. Solid arrow on bottom indicates when ecopassages were installed in 2017; broken arrow
indicates the installation of drift fencing prior to the 2019 active season.
35

0.35
0.30

50

2015 and 2016

A

N=377

C

2015 and 2016

D

2019 and 2020

N=92

40

0.25

30

0.15
0.10
0.05
0.00
0.30

2019 and 2020

B

N=490

0.25
0.20
0.15

Roadkilled rattlesnakes

Proportion of captures

0.20

20
10
0
20

N=55

15
10

0.10

5
0.05
0.00

- 80

- 90

70

80

10

- 70
60

0

- 60
50

0- 1

- 50
40

- 10

- 40
30

Snout-Vent Length (cm)

Neonate Juvenile Subadult

Adult

10

- 30
20

90

- 20
10

0

Age class

Figure 2.2. A & B) Proportion of Western Rattlesnakes captured across size classes at six focal dens in the White Lake Basin before
(A) and after (B) ecopassages were installed. Snakes with measured lengths falling exactly on an interval cut-off were included in the
larger category. C & D) Number of roadkilled Western Rattlesnakes in broad age categories as detected in the White Lake Basin
before (C) and after (D) ecopassages were installed. Neonates are defined as rattlesnakes with snout-vent lengths of < 30 cm, juveniles
are 30-40 cm, subadults are 40-55 cm, and adults are > 55 cm.
36

37

Figure 2.3. Jolly-Seber estimates (± 1 SE) for population size and survival of all (top) and
adult (bottom) Western Rattlesnakes (Crotalus oreganus) at focal dens in the White Lake
Basin, BC from 2015-2020. Also shown in the top graph is the original Winton et al. (2020)
population estimate for 2016 (■) as calculated using the shorter mark/recapture history
available during their study. Linear regression lines for the population estimates of all snakes
(y = 68.3x + 488.1, R2 = 0.33, P = 0.32) and adults snakes (y = 4.9x + 309.3, R2 = 0.02,
P = 0.81) are also shown. The solid arrow represents the installation of snake-targeted
ecopassages and the dashed arrow indicates the installation of directional fencing as road
mortality mitigation.

38
follow scent trails of conspecifics to locate both mates and hibernacula (Costanzo 1989,
LeMaster et al. 2001). Thus, increased usage may even be more exponential than linear.
Given the duration of this study is considerably less than the estimated generation time for
Western Rattlesnakes in British Columbia, this suggests that a numerical response of the
population would take more time than the two years post-mitigation period to show clear
positive effects. However, at this point there is insufficient data to determine how long the
‘shock phase’ will persist, or if this species will exhibit different recovery phases. Before
longer-term data are available, the conservation Precautionary Principle should be invoked,
namely that “when an activity raises threats of harm to human health or the environment,
precautionary measures should be taken even if some cause and effect relationships are not
fully established scientifically” (Raffensperger and Tickner 1999). However, this must be
balanced against the fact that the deployment of ecopassages is costly, both in terms of
installation and maintenance.
Mark-recapture success, and the subsequent confidence in Jolly-Seber estimates, is
affected by a number of factors under varying levels of control by the researchers. Changes
in field personnel in 2021 likely affected survey effort, as will weather patterns and other
seasonal variants that influence emergence patterns and capture success of snakes. These
factors may have contributed to the differences in population estimates I calculated across the
latter years of my study, but this non-conclusive outcome would be hard to eliminate given
the nature of this study. I believe the main conclusions made herein remain sound.
Understanding the rate at which mitigation efforts, including ecopassages and
fencing, generate behavioural and numerical responses in smaller vertebrates remains
important for assessing their efficacy. As with natural disturbances, immediate responses
may differ considerably from longer-term ones, but together they paint a complete picture of
the response which can aid in the conservation of populations. This study reveals a potential
decline in Western Rattlesnake roadkill in the ‘shock phase’ immediately following
ecopassage installation, although longer-term work beyond the ‘shock-phase’ is needed. This
study shows that immediate, obvious effects from mitigation efforts on target populations
should not be expected, while providing a cogent argument for ensuring commitment to
projects like this are made for the long term.

39

LITERATURE CITED
Altwegg R, Dummermuth S, Anholt BR, Flatt T. 2005. Winter weather affects Asp Viper
Vipera aspis population dynamics through susceptible juveniles. Oikos. 110: 55-66.
Andrews KM, Gibbons JW. 2005. How do highways influence snake movement? Behavioral
responses to roads and vehicles. Copeia. 2005(4): 772-782.
Baillargeon S, Rivest L-P. 2007. Rcapture : Loglinear Models for capture-recapture in R.
Journal of Statistical Software. 19: 1–31.
Baxter-Gilbert JH, Riley JL, Lesbarrères D, Litzgus JD. 2015. Mitigating reptile road
mortality: Fence failures compromise ecopassage effectiveness. PLoS ONE 10(3).
Boyle SP, Keevil MG, Litzgus JD, Tyerman D, Lasbarréres D. 2021. Road-effect mitigation
promotes connectivity and reduces mortality at the population-level. Biological
Conservation. 261.
Brown WS, MacLean FM. 1983. Conspecific scent-trailing by newborn Timber
Rattlesnakes, Crotalus horridus. Herpetologica. 39(4): 430-436.
Colley M, Lougheed SC, Otterbein K, Litzgus JD. 2017. Mitigation reduces road mortality of
a threatened rattlesnake. Wildlife Research. 44(1): 48-59.
Costanzo JP. 1989. Conspecific scent trailing by garter snakes (Thamnophis sirtalis) during
autumn: Further evidence for use of pheromones in den location. Journal of Chemical
Ecology. 15(11): 2531-2538.
Dillon RM, Boyle SP, Litzgus JD, Lesbarrères D. 2020. Build it and some will use it: A test
of road ecopassages for Eastern Gartersnakes. Journal of Herpetology. 54(1): 19–23.
Duvall D, Goode MJ, Hayes WK, Leonhardt JK, Brown DG. 1990. Prairie Rattlesnake
vernal migration: Field experimental analyses and survival value. National
Geographic Research. 6(4): 457-469.
Fahrig L, Rytwinski T. 2009. Effects of roads on animal abundance: An empirical review and
synthesis. Ecology and Society. 14(1): 21.
Ford AT, Barrueto M, Clevenger AP. 2017. Road mitigation is a demographic filter for
Grizzly Bears. Wildlife Society Bulletin. 41(4): 712-719.
Forman RT, Sperling TD, Bissonette JA, Clevenger AP, Cutshall CD, Dale VH, Fahrig L,
France R, Goldman CR, Heanue K, et al. 2003. Road Ecology: Science and Solutions.
Washington (DC): Island Press. 481 p.

40
Gilhooly PS, Nielsen SE, Whittington J, St. Clair CC. 2019. Wildlife mortality on roads and
railways following highway mitigation. Ecosphere. 10(2): e02597.
Gomez L, Larsen KW, Gregory PT. 2015. Contracting patterns of migration and habitat use
in neighboring rattlesnake populations. Journal of Herpetology. 49(3): 371-376.
Jacobson SL, Bliss-Ketchum LL, de Rivera CE, Smith WP. 2016. A behavior-based
framework for assessing barrier effects to wildlife from vehicle traffic volume.
Ecosphere. 7(4).
Jarvis LE, Hartup M, Petrovan SO. 2019. Road mitigation using tunnels and fences promotes
site connectivity and population expansion for a protected amphibian. European
Journal of Wildlife Research. 65: 27.
Krebs CJ. 1989. Ecological Methodology. New York (NY): HarperCollinsPublishers, Inc.
654 p.
LeMaster MP, Moore IT, Mason RT. 2001. Conspecific trailing behaviour of red-sided garter
snakes, Thamnophis sirtalis parietalis, in the natural environment. Animal Behaviour.
61: 827-833.
Lomas E, Maida JR, Bishop CA, Larsen KW. 2019. Movement ecology of Northern Pacific
Rattlesnakes (Crotalus o. oreganus) in response to disturbance. Herpetologica. 75(2):
153-161.
Loraamm R, Anderson J, Burch C. 2021. Identifying road avoidance behaviour using timegeography for Red Deer in Banff National Park, Alberta, Canada. Transactions in
GIS. 2021(25): 1331-1346.
Macartney JM. 1985. The ecology of the Northern Pacific Rattlesnake, Crotalus viridis
oreganus, in British Columbia [thesis]. Victoria, BC, Canada: University of Victoria.
Macartney JM, Gregory PT. 1988. Reproductive biology of female rattlesnakes (Crotalus
viridis) in British Columbia. Copeia. 1988(1): 47–57.
Macartney JM, Gregory PT, Charland MB. 1990. Growth and sexual maturity of the Western
Rattlesnake, Crotalus viridis, in British Columbia. Copeia. 1990(2): 528-542.
Madsen T, Ujvari B, Shine R, Olsson M. 2006. Rain, rats and pythons: Climate-driven
population dynamics of predators and prey in tropical Australia. Austral Ecology.
31(1): 30-37.
Maida JR, Kirk DA, McKibbin O, Row JR, Larsen KW, Stringham C, Bishop CA. 2018.
Population estimate, survivorship, and generation time of the Northern Pacific
Rattlesnake (Crotalus o. oreganus) at its northern-most range limits. Herpetological
Conservation and Biology. 13(3): 662-672.

41

Meidinger D, Pojar J. 1991. Ecosystems of British Columbia. Victoria, British Columbia,
Canada: British Columbia Ministry of Forests. Available from:
https://www.for.gov.bc.ca/hfd/pubs/Docs/Srs/Srs06.pdf
Muellman PJ, Da Cunha O, Montgomery CE. 2018. Crotalus horridus (Timber Rattlesnake)
maternal scent trailing by neonates. Northeastern Naturalist. 25(1): 50-55.
Nowak EM, Theimer TC, Schuett GW. 2008. Functional and numerical responses of
predators: Where do pit vipers fit in the traditional paradigms? Biological Reviews.
83: 601-620.
Parent C, Weatherhead PJ. 2000. Behavioral and life history responses of Eastern
Massasauga Rattlesnakes (Sistrurus catenatus catenatus) to human disturbance.
Oecologia. 125: 170-178.
Parker JM, Anderson SH. 2007. Ecology and behavior of the Midget Faded Rattlesnake
(Crotalus oreganus concolor) in Wyoming. Journal of Herpetology. 41(1): 41-51.
Paterson JE, Baxter-Gilbert J, Beaudry F, Carstairs S, Chow-Fraser P, Edge CB, Lentini AM,
Litzgus JD, Markle CE, McKeown K, et al. 2019. Road avoidance and its energetic
consequences for reptiles. Ecology and Evolution. 2019;00: 1-10.
Raffensperger C, Tickner J, editors. 1999. Protecting public health and the environment:
Implementing the Precautionary Principle. Washington (DC): Island Press. 353 p.
Rajaonson J, Tanguay GA. 2021. The COVID-19 pandemic has created regional tourism
hotspots as big cities suffer. The Conversation (Canada). Available from:
https://theconversation.com/the-covid-19-pandemic-has-created-regional-tourismhotspots-as-big-cities-suffer-159892 [accessed 2022 Nov 10].
Rytwinski T, Soanes K, Jaeger JAG, Fahrig L, Findlay CS, Houlahan J, van der Ree R, van
der Grift EA. 2016. How effective is road mitigation at reducing road-kill? A metaanalysis. PLoS ONE. 11(11).
Schmidt DA, Govindarajulu P, Larsen, KW, Russello MA. 2020. Genotyping-in-thousands
by sequencing reveals marked population structure in Western Rattlesnakes to inform
conservation status. Ecology and Evolution. 10(14): 7157-7172.
Scudder KM, Chiszar D, Smith HM, Melcer T. 1988. Response of neonatal Prairie
Rattlesnakes (Crotalus viridis) to conspecific and heterospecific chemical cues. The
Psychological Record. 38: 459-471.

42
Seidler RG, Green DS, Beckman JP. 2018. Highways, crossing structures and risk: Behaviors
of Greater Yellowstone pronghorn elucidate efficacy of road mitigation. Global
Ecology and Conservation. 15 (2018).
Shine R, Brown GP, Goiran C. 2021. Population dynamics of the sea snake Emydocephalus
annulatus (Elapidae, Hydrophiinae). Scientific Reports. 2021(11): 20701.
The Nature Trust of British Columbia [TNTBC]. c2022. Vancouver, BC; [accessed 2022
Aug 2]. https://www.naturetrust.bc.ca/news/legacy-landscapes-white-lake-basinbiodiversity-ranch .
Warren SD, Scifres CJ, Teel PD. 1987. Response of grassland arthropods to burning: A
review. Agriculture, Ecosystems & Environment. 19:105-130.
Whelan RJ, Rodgerson L, Dickman CR, Sutherland EF. 2002. Critical life cycles of plants
and animals: Developing a process-based understanding of population changes in
fire-prone landscapes. In: Bradstock RA, Williams JE, Gill MA, editors. Flammable
Australia: The Fire Regimes and Biodiversity of a Continent. Cambridge (UK):
Cambridge University Press. p. 94-104.
Winton SA. 2017. Road mortality hot spots and movement corridors of at-risk snake species
in the White Lake Basin of British Columbia. Prepared for the BC Ministry of
Environment. Unpublished.
Winton SA. 2018. Impacts of road mortality on the Western Rattlesnakes (Crotalus
oreganus) in British Columbia [MSc thesis]. Kamloops, BC, Canada: Thompson
Rivers University.
Winton SA, Bishop CA, Larsen KW. 2020. When protected areas are not enough: Lowtraffic roads projected to cause a decline in a northern viper population. Endangered
Species Research. 41:131–139.
Winton SA, Taylor R, Bishop CA, Larsen KW. 2018. Estimating actual versus detected road
mortality rates for a northern viper. Global Ecology and Conservation. 16 (2018).
Wong C. In prep. Movement ecology and road mortality for a community of threatened
snakes in British Columbia, Canada [MSc thesis]. Kamloops, BC, Canada: Thompson
Rivers University.

CHAPTER 3
DIFFERENT SNAKE SPECIES SHOW VARYING RESPONSES TO ECOPASSAGE
INSTALLATION

INTRODUCTION

Crossing structures often are used to lessen the impacts of roads on wildlife, as they
reduce roadkill and aid in habitat connectivity (Glista et al. 2009, Jarvis et al. 2019, van der
Grift et al. 2013). There has been extensive research done on their efficacy for large
mammals, since collisions with such wildlife have safety and economic impacts on humans
(Donaldson and Elliott 2021, Lee et al. 2020, Rytwinski et al. 2016). However, smaller
animals face the same road effects and, although collisions with vehicles cause minimal
damage to humans, these species can benefit from mitigation structures just the same (Fahrig
and Rytwinski 2009, McGregor et al. 2007, Plante et al. 2019).
Reptiles are one of the fastest declining taxa globally, and particularly vulnerable to
road effects due to their small size, attraction to roads for thermoregulation, and low vehicle
avoidance (Baxter-Gilbert et al. 2015, Fahrig and Rytwinski 2009). The implementation of
mitigation measures to reduce road mortality while improving habitat connectivity for
vulnerable reptile species has become increasingly common (Jarvis et al. 2019, Dillon et al.
2020). To facilitate the crossing of road corridors, ‘ecopassages’ are installed under roads
and coupled with barrier fencing designed to direct wildlife into the passageways. Such
initiatives often are prompted by a recognition or concern that roads are threatening the
viability of wildlife populations (Aresco 2005, Colley et al. 2017, Polak et al. 2019).
Although the application of these mitigative tactics is increasing, few studies have fully
assessed their use and resulting effect on road mortality for the target species (van der Grift et
al. 2013, Baxter-Gilbert et al. 2015).
Intuitively, one of the most efficient and cost-effective methods of monitoring
crossing structures are wildlife cameras (Ford et al. 2009, Welbourne et al. 2017). Cameras
may be installed inside ecopassages and set to take movement-activated photos and/or timed
43

44
photos at selected intervals, enabling a considerable amount of data to be collected with
minimal effort, and without disturbing the animals (Wearn and Glover-Kapfer 2019). Photos
from wildlife cameras provide information on species’ use, and allow comparisons between
species (Pagnucco et al. 2011, Wearn and Glover-Kapfer 2019). Although research using
camera traps has primarily been focused on mammals, fish, and birds, with improving
technology it can be a useful monitoring tool for reptiles and amphibians as well (Molyneux
et al. 2017, Wearn and Glover-Kapfer 2019, Welbourne et al. 2017).
Ecopassages for snakes were established in southern British Columbia, Canada, after
research (Winton et al. 2018, 2020) documented significant road mortality and a population
decline in Western Rattlesnakes (Crotalus oreganus). Cameras were used to monitor wildlife
use, principally the three at-risk snake species in the area: Western Rattlesnakes (F.
Viperidae - Crotalus oreganus), and two species in the Family Colubridae: Great Basin
Gophersnakes (Pituophis catenifer deserticola), and Western Yellow-bellied Racers
(Coluber constrictor mormon). In addition, drift fences were installed at ecopassage
entrances prior to the snakes’ active season in 2019 (see Chapter 1). In this chapter I compare
camera detections in the ecopassages between the three snake species immediately upon their
installation (the ‘shock phase’ – See Chapter 2).
I predicted that the general response to the installation of the new ecopassages and
drift fencing would vary among the three target species in the White Lake Basin. Despite
occupying the same ecosystem, the three species differ in habitat requirements and foraging
strategies. Rattlesnakes leave communal hibernacula on rocky, warm slopes in the spring for
summer foraging grounds, where they spend much of their time under cover of rocks and
shrubs. They are ambush predators (Clark 2004, Theodoratus and Chiszar 2000), largely
remaining concealed so that they can forage successfully (B.C. Ministry of Water, Land and
Air Protection 2004c, COSEWIC 2015). In contrast, gophersnakes and racers appear to use a
broader range of locations for hibernating and are active foragers (B.C. Ministry of Water,
Land and Air Protection 2004a and 2004b). Gophersnakes search rodent holes and climb to
reach bird nests, and racers will approach and chase their prey (B.C. Ministry of Water, Land
and Air Protection 2004a and 2004b, COSEWIC 2013). Because of this, these two members

45
of the colubrid family of snakes move more often and with more tortuosity than rattlesnakes
(Rouse 2006, Wong in prep).
I predicted that ecopassage use would increase over time for all three species of
snakes. This trend has been observed in large mammals, where the animals become
acclimatized and learn to adopt the structures (Ford et al. 2017, Gilhooly et al. 2019, Seidler
et al. 2018). In addition, snakes rely on olfactory cues for navigation, and scent trails from
conspecifics develop over time (Brown and MacLean 1983, Costanzo 1989, Muellman et al.
2018). Although the timeframe of my project was relatively short, it focused on a very
important period when an assessment of mitigation success is highly desirable, particularly if
other similar work is being contemplated. All three of my target species are listed as
Threatened under the Canadian Species at Risk Act (2002) and recognized as ‘Special
Concern’ within the province of British Columbia (Racer Management Team Working
Group 2013, Southern Interior Reptile and Amphibian Working Group 2016a and 2016b).
Thus, the two main objectives of my study were to (1) quantify the short-term response of
these three species to newly installed mitigation structures, and (2) compare differences in
temporal and spatial ecopassage use between species. A secondary objective of this study
was to assess the immediate efficacy of drift fence associated with ecopassages.

METHODS
Study Site
This study was conducted in the White Lake Basin (LAT 49.318N, LONG
119.638W) in the South Okanagan region of British Columbia, Canada. As suggested
above, research at this site has been taking place since 2015, providing pre-mitigation
roadkill data. This location lies near the northern limit of the three target species’ respective
ranges (Fig. 1, Winton et al. 2018), and the annual activity period of the snakes here
generally extends from April – October of each year. The basin consists of open shrub-steppe
grassland habitat surrounded by rolling hills and steep bluffs, with two golf courses and small
residential communities and ponds lying just outside the basin. Two roads – one running

46
largely East-West (White Lake Road) and one running North-South (Willowbrook Road) –
meet at a T-intersection (Figure 3.1).
In September 2017, 4 ecopassages were installed throughout the study area at
locations where snake roadkill occurrences were the highest in 2015-2016 (Winton unpubl.).
In April 2019 drift fencing was added to 3 of these new ecopassages and 4 existing drainage
culverts to funnel snakes into the openings (Figure 1.3). In total, eight ecopassages were
created with the specific intent of lowering rattlesnake road mortality within a 6.5 km stretch
of road (Figure 3.1). Some of the ecopassages are in close proximity to known rattlesnake
dens while others lie further away.
Ecopassages were made of corrugated metal, with an average height of 45 ± 8 SD cm
and an average width of 67 ± 12 SD cm, with an average openness of 0.025 ± 0.005 SD m (=
width x height / length; Clevenger and Waltho 2001). The ecopassages that were fully new
installations (n=4) were more oval-shaped than the rounder drainage culverts that were
retained and refurbished (Figure 1.4. A & B). A substrate of sand (≈ 5 cm deep) lined the
bottom of all eight ecopassages.
Drift fencing is recommended to increase ecopassage use by reptiles (Fahrig and
Rytwinski 2009, Forman et al. 2003), so in April 2019 drift fencing was added to the
entrances of the ecopassages to putatively increase snake usage (Figures 1.3 and 2.1).
Bedrock, drainage patterns (including snowmelt), potential danger to cyclists, and cattle use
affected a controlled, uniform experimental design for the drift fences. As a result, 7 drift
fences were established, with only one fence being situated at a single entrance to each of 7
ecopassages. Drift fencing was not installed at one ecopassage because the surrounding
terrain was deemed too steep and rocky. The drift fencing was made of black, 2mm-thick
recycled High Density Polyethylene (model AMX-SP40, Animex Wildlife Fencing
Solutions), and the top of the drift fences were folded over away from the road to create a lip,
making it more difficult for snakes to climb over (See Figure 1.4. B, C & D). Fences were
~75cm in height, although this varied (± 5 cm) within and between fences, depending on the
depth of the substrate around the base of the fence section. When constructed, the fence
lengths flanking the entrances of the ecopassages had an average length of 52 ± 8.3 m,

47
representing 0.4 km out of 23.4 km of road edge within the study site (Fig. 1, Winton et al.
2018).

Camera Monitoring
I used Bushnell HD Natureview cameras (model numbers 119439, 119440, and
119740) equipped with 46 cm focal lenses to monitor snake use of the ecopassages. The
cameras were located immediately inside the entrances of all eight ecopassages, resulting in a
total of 16 deployed cameras. Each camera was mounted facing inwards using a lock box
secured to either the top or side of the ecopassage with a large bolt screw (See Figure 1.4. A
& B). Two trigger methods were used to secure pictures inside the culverts: the cameras were
set to high sensitivity and took 3 consecutive photos (each 1 second apart) when activated by
movement, as well as timed captures each hour on the hour. The cameras were active from
early April to early October of 2019 and 2020. Every two weeks the Secure Digital (SD)
cards were switched out and the batteries replaced if necessary. Photos were manually
processed to identify the presence of snakes. Data recorded for each animal captured on
camera included date, time, species, culvert entrance ID, field scan/ motion sensor activation,
direction of travel, and whether the appearance resulted in a passage (see below).
An appearance was recorded any time a snake was detected in an ecopassage. In
some instances, an individual was captured on camera multiple times within one appearance
event. For example, if a snake remained coiled in the entrance of an ecopassage for multiple
hours, it was only considered one appearance despite being photographed multiple times
throughout its stay. If an image of a snake was linked to a passage (e.g. photographed twice,
once at each end), then that snake was only considered as one appearance. (e.g. total
detections – passages = appearances).
Photo quality generally made it impossible to be 100% confident that the same
individual was entering and exiting an ecopassage. However, the infrequent passage of
snakes (especially within species) allowed me to designate a passage as having occurred
when a snake of the same species and appearance (size, skin patterns, etc.) was photographed
entering and exiting opposite ends of an ecopassage within a 30-minute time interval.

48

Figure 3.1. Map of camera-monitored ecopassage locations on White Lake and Willowbrook
Roads, in the White Lake Basin, British Columbia, Canada. Ecopassages labelled in red are
existing, modified drainage culverts, and ecopassages labelled in yellow were newly-created
in September 2017 at snake roadkill hotspots. Circles indicate the side of the road that drift
fencing was on. Lines representing ecopassages are not drawn to scale. (Base map from
Google Earth)

49
An entrance was recorded when a snake was photographed moving inwards from an
ecopassage opening. This metric was necessary to assess the efficacy of the drift fences.
Entrances from a fenced end of an ecopassage indicate that a fence may have aided in
directing a snake towards an opening, whereas exits from a fenced end would not.

Data Analysis
Paired T-tests were used to compare entrances of snakes through an ecopassage
opening with a drift fence versus the opposing opening lacking a drift fence, with the
ecopassage being the basis of the pairing. I did 6 tests in total, quantifying differences within
species (rattlesnakes, racers, gophersnakes) for both years of the study (2019 and 2020).
I used a χ2 test to determine if ecopassage preference differed between species by
comparing the proportion of appearances of each species in each ecopassage.

RESULTS
In each of my two study years, I recorded over 1000 appearances of racers, whereas
gophersnakes and rattlesnakes provided less than 110 appearances each (Table 3.1). Passages
for all species were far less common, but I detected over 300 racer passages in both years
while gophersnakes and rattlesnakes produced less than 25 in either year. However, in 2020
all three species increased in appearances and passages (Table 3.1). Overall, gophersnakes
showed slightly more ecopassage use (appearances and passages) than rattlesnakes, and
racers showed substantially more use than either of the other two species. Of the two metrics,
I opted to use appearance data for the χ2 test comparing ecopassage preference since there
was significantly less passage data for each species (Table 3.1).
The three species differed in their ecopassage use both temporally and spatially.
There was an obvious lack of any temporal patterns both among species and between years,
apart from rattlesnake use peaking mid-season in both 2019 and 2020 (Figure 3.2). The
relative use of the different ecopassages was strikingly different across species (Figure 3.3 -

50
χ2 = 497.3, df = 14, P < 0.001), with rattlesnakes, gophersnakes, and racers most frequently
being detected in ecopassages 1, 8, and 4, respectively.
The only instance in which the frequency of entrances through ecopassage ends with
drift fences and non-fenced ends differed significantly was with rattlesnakes in 2020, in
which they entered through fenced ends more often (Table 3.2). However, the percent of
entrances through fenced ends increased for all three species from 2019 to 2020 (Table 3.2).
Both rattlesnakes and gophersnakes entered a new ecopassage most frequently, whereas for
racers the four ecopassages with the most entrances all were the modified ones (Fig 3.4).

DISCUSSION

As predicted, there was substantial variation among the species in their use of the
ecopassages quantitatively, temporally, and spatially. Nonetheless, snakes of all three species
were found to be transiting the ecopassages within the two years immediately after their
installation. This is an early sign of the ecopassage effects, particularly if one assumes each
time a snake uses an ecopassage it avoids a road surface crossing. Of interest is the fact that
rattlesnakes used the ecopassages the least of the three at-risk snake species, while racers
used them considerably more often than the other two snake species. I cannot conclude how
these numbers relate to the relative abundance of the three species: although population sizes
and densities are available for the rattlesnakes in this area (owing largely to their denning
behaviour), comparable data for gophersnakes and racers are much more difficult to obtain.
However, the overwhelming disproportionate use by racers relative to the other two species
suggest that these species are showing markedly different responses to the ecopassages.
Despite major variation in temporal use of the ecopassages throughout the active
seasons, one interesting commonality is that in both 2019 and 2020 the peak of rattlesnake
ecopassage use was around the midway point of the summer. In British Columbia
rattlesnakes migrate away from their dens in the spring (April-June) and back in the fall
(September-October) (Macartney 1985, Maida et al. 2020), so those times would be when

51

Table 3.1. Ecopassage camera appearances and passages of three ‘at-risk’ snake species in 8
ecopassages in the White Lake Basin, British Columbia, Canada. Percentages show how
many appearances resulted in passages (Passages/Appearances*100). By way of example, in
2019 there were 53 appearances recorded for rattlesnakes, and of those, only 3 (or 6%)
resulted in a documented passage. See text for working definitions of ‘appearances’ and
‘passages’.

Appearances

Passages

Year

2019

2020

2019

%

2020

%

C. oreganus

53

62

3

6

9

15

P. c. deserticola

74

109

7

9

22

20

C. c. mormon

1219

1842

346

28

402

22

Table 3.2. Ecopassage entrances of three ‘at-risk’ snake species detected by wildlife cameras in 7 ecopassages in the White Lake
Basin, British Columbia, Canada. One end of each ecopassage is flanked by drift fencing, whereas the opposite entrance is not. Paired
T-tests were used to compare the number of fenced vs. non-fenced entrances for each species in each of the two years of the study, and
the resulting P values are displayed here, with an asterisk (*) denoting a significant difference.

Entrances through
fenced ends

Entrances through
non-fenced ends

% Entrances through
fenced ends

P

Year

2019

2020

2019

2020

2019

2020

2019

2020

C. oreganus

12

22

10

12

55

65

0.680

0.025*

P. c. deserticola

17

31

22

19

44

62

0.140

0.439

C. c. mormon

295

488

422

482

41

50

0.196

0.977

52

53

14

200

Rattlesnakes
Gophersnakes
Racers

2019

180

12

160
140
120

8

100

6

80
60

4

40
2

20

0
14

0
200

2020

180

12

Appearances (Racers)

Appearances (Rattlesnakes and Gophersnakes)

10

160
140

10

120
8

100

6

80
60

4

40
2

20

0

0
1

3

5

7

9

11

13

15

17

19

21

23

25

Week
Figure 3.2. Weekly ecopassage (n=8 ecopassages) camera appearances of Western
Rattlesnakes, Great Basin Gophersnakes, and Western Yellow-bellied Racers from April 15
to October 6, 2019 and 2020 in the White Lake Basin of British Columbia, Canada.

54

0.35

Rattlesnakes (n=115)

0.30
0.25
0.20
0.15
0.10
0.05

Proportion of Appearances

0.00
0.35

Gophersnakes (n=183)

0.30
0.25
0.20
0.15
0.10
0.05
0.00
0.35

Racers (n=3061)

0.30
0.25
0.20
0.15
0.10
0.05
0.00
1

2

3

4

5

6

7

8

Ecopassage
Figure 3.3. Proportion of appearances of Western Rattlesnakes, Great Basin Gophersnakes,
and Western Yellow-bellied Racers in 8 ecopassages in the White Lake Basin, British
Columbia, Canada in 2019 and 2020 combined. Dark-coloured bars represent newly installed
ecopassages, and light-coloured bars represent pre-existing culverts that were modified with
drift fencing. See Figure 3.1 for ecopassage locations.

55

10

Rattlesnake

8

New
Modified

6
4
2

Unfenced Entrance

0
14

0

2

4

6

8

10 12 14 16 18

Gophersnake

12
10
8
6
4
2
0
0
350

2

4

6

8

10 12 14 16 18

Racer

300
250
200
150
100
50
0
0

50 100 150 200 250 300 350

Fenced Entrance
Figure 3.4. Snake entrances through drift fenced versus unfenced ends of seven ecopassages
in 2019 and 2020 combined. The dotted line represents hypothetical equal entrances from
both ends. Closed circles represent newly installed ecopassages, and open circles represent
pre-existing culverts that were modified with drift fencing.

56
snakes are most likely to be detected in association with ecopassages and/or roads. However,
the ecopassage camera data did not demonstrate strong shoulder-season spikes in activity.
Perhaps rattlesnakes are attracted to the road for thermoregulatory reasons or continue
moving mid-season to find mates or food (Harvey and Larsen 2020). Conversely,
gophersnakes and racers appeared to lack consistency in temporal usage patterns between
years. Since the denning and egg-laying locations of racers and gophersnakes are more
variable than that of rattlesnakes (B.C. Ministry of Water, Land and Air Protection 2004a,
2004b and 2004c, McKelvey in prep) individuals of these species may spend more of the
active season near the ecopassages, and therefore encounter them more frequently and
sporadically.
The spatial dissimilarities in ecopassage use among these species likely reflect, to
some degree, differences in habitat selection, although without more detailed study this
statement is largely conjecture. Still, the ecopassage with the most rattlesnake detections lied
relatively close to two rattlesnake communal hibernacula on a roadkill hotspot corresponding
to a migratory corridor. Conversely, the ecopassages used most often by racers were in close
proximity to a pond offering, relatively moist habitat in a near-desert ecosystem. Racers are
highly insectivorous and forage in areas of abundant herbaceous vegetation, like riparian
habitat (Fleet et al. 2009). Ideally, the number and placement of ecopassages would stratify
habitat believed important to a multitude of species, but admittedly this may be a lofty and
expensive task.
The increase in detection rates of snakes in ecopassages from 2019 to 2020 is
consistent with the notion that snakes are particularly responsive to olfactory cues (see
below), allowing them to build up familiarity with new passageways (Brown and MacLean
1983, Costanzo 1989, Muellman et al. 2018). It is also consistent with a reduction in
rattlesnake roadkill rates in the area from 2019 to 2020 (Chapter 2). However, without longer
term data, it would be premature to conclude this pattern was at work.
Similarly, snake entrances through fenced ends of the ecopassages also increased
from 2019 to 2020. The year of drift fence installation (2019) showed a muted immediate
response by the snakes in terms of fenced versus unfenced entrances to culverts. However,
the following year (2020), the percentage of entrances through fenced ecopassage entrances

57
increased for all three species, and the number of rattlesnakes entering fenced ends was
significantly higher than the number at non-fenced ends. Previous research at other locations,
on other species, suggests that road mortality would be reduced by expanding the fencing
(Baxter-Gilbert et al. 2015, Colley et al. 2017, Markle et al. 2017, Plante et al. 2019).
Although installation constraints impacted my research design, in one year of this study
rattlesnake ecopassage use was indeed higher according to the presence of fencing. Currently
the fences surrounding the culvert entrances provide relatively short, inconsistent barriers to
snake movement within the study area. Intuitively, their overall impact on snake road
mortality would be enhanced if the fences were expanded.
Olfactory cues and conspecific scent trailing are important for snakes, particularly
communally aggregating snakes like rattlesnakes (Brown and MacLean 1983, Muellman et
al. 2018). Thus, ecopassage appearances and passages likely will continue to increase as
scent trails develop. The impact of increased familiarity/scenting may be linked to the
relatively higher entrance counts for pre-existing, modified ecopassages than the newly
installed ones, most notably for racers. Possibly racers rely more heavily on scent trails, or
their movement patterns result in repeated trips more often in the same area (Wong in prep)
In the case of rattlesnakes and gophersnakes, there were obvious outlying ecopassages (both
newly-installed) that experienced more entrances both in general and through fenced ends.
This may be because these were ideal locations for the ecology of that species. The
ecopassage experiencing the most rattlesnake entrances has a known den on either side of the
road, so in this case the number of snakes nearby and/or the ideal location of the ecopassage
likely outweighs the unfamiliarity of it.
Although the timeframe encompassed in this study is relatively short, this initial
period following the installation of mitigation measures provides an important snapshot of
the immediate response of snakes to these structures. In Chapter 2, I likened this period to
that described by Warren et al. (1987) and Whelan et al. (2002) for the recovery of
ecosystems and species immediately after a wildfire (the so-called ‘shock phase’).
Conducting assessments of mitigation effects through this time period will prove valuable for
understanding long-term patterns (which also should be monitored), including adding to our
knowledge of how quickly target species respond to new structures of this type. In this study,

58
I have shown clearly that even three closely-related species respond differently to the same
crossing structures, although they all used ecopassages to cross roads, presumably in lieu of
crossing on the road’s surface. Differences in crossing structure use of similar taxa has been
reported elsewhere, for example mule deer more readily use underpasses, and in general use
crossing structures twice as often as sympatric pronghorn, which prefer overpasses (Sawyer
et al. 2016). In this case, snake crossing varied by frequency, time, and location among
species. To accommodate this variation and direct more efficient use of conservation funds,
ecopassages intended for multi-taxa use may be installed (Polak et al. 2019, Santini et al.
2016). Photos from the cameras at this study site have shown at least 31 other species using
the ecopassages – many being of conservation concern in British Columbia - including the
American Badger (Taxidea taxus jeffersonii) and Western Tiger Salamander (Ambystoma
mavortium) (Matson 2021, Appendix D). Both of these species often fall victim to roadkill
(Crosby 2014, Sunga et al. 2017). Facilitating the use of ecopassages by a wide range of
species should also be considered when designing fencing along the entrances (Aresco 2005,
Baxter-Gilbert et al. 2015, Cunnington et al. 2014).
It is important to recognize that although all three snake species in this study used the
ecopassages to cross roads, this metric alone cannot be used to conclude the mitigation
efforts are having a significant effect. Ideally work of this nature needs to be combined with
demographic and roadkill monitoring (as done for rattlesnakes in Chapter 2). In particular,
the short timeframe of this study, and lack of population data for racers and gophersnakes,
makes it premature to conclude on the merits of the mitigation work. But, in the absence of
noticeable negative effects, the Precautionary Principle and the continued application of these
techniques appears warranted.

59

LITERATURE CITED
Anđalković M, Bogdanović N. 2022. Amphibian and reptile road mortality in special nature
reserve Obedska Bara, Serbia. Animals. 12(561).
Aresco MJ. 2005. Mitigation measures to reduce highway mortality of turtles and other
herpetofauna at a north Florida lake. Journal of Wildlife Management. 69(2): 549560.
Baxter-Gilbert JH, Riley JL, Lesbarrères D, Litzgus JD. 2015. Mitigating reptile road
mortality: Fence failures compromise ecopassage effectiveness. PLoS ONE 10(3).
B.C. Ministry of Water, Land and Air Protection. 2004a. Great Basin Gopher Snake
(Pituophis catenifer deserticola) in Accounts and Measures for Managing Identified
Wildlife – Accounts V. 2004. B.C. Ministry of Water, Land and Air Protection,
Victoria, B.C. Available from: https://www.env.gov.bc.ca/wld/frpa/iwms/documents/
Reptiles/r_greatbasingophersnake.pdf
B.C. Ministry of Water, Land and Air Protection. 2004b. Racer (Coluber constrictor
mormon) in Accounts and Measures for Managing Identified Wildlife – Accounts V.
2004. B.C. Ministry of Water, Land and Air Protection, Victoria, B.C. Available
from: https://www.env.gov.bc.ca/wld/frpa/iwms/documents/Reptiles/r_racer.pdf
B.C. Ministry of Water, Land and Air Protection. 2004c. Western Rattlesnake (Crotalus
oreganus) in Accounts and Measures for Managing Identified Wildlife – Accounts V.
2004. B.C. Ministry of Water, Land and Air Protection, Victoria, B.C. Available
from: https://www.env.gov.bc.ca/wld/frpa/iwms/documents/Reptiles/r_westernrattle
snake.pdf
Brown WS, MacLean FM. 1983. Conspecific Scent-trailing by newborn Timber
Rattlesnakes, Crotalus horridus. Herpetologica. 39(4): 430-436.
Clark RW. 2004. Timber Rattlesnakes (Crotalus horridus) use chemical cues to select
ambush sites. Journal of Chemical Ecology. 30(3): 607-617.
Clevenger AP, Waltho N. 2001. Factors influencing the effectiveness of wildlife underpasses
in Banff National Park, Alberta, Canada. Conservation Biology. 14(1): 47-56.
Colley M, Lougheed SC, Otterbein K, Litzgus JD. 2017. Mitigation reduces road mortality of
a threatened rattlesnake. Wildlife Research. 44(1): 48-59.
Costanzo JP. 1989. Conspecific scent trailing by garter snakes (Thamnophis sirtalis) during
autumn further evidence for use of pheromones in den location. Journal of Chemical
Ecology. 15(11): 2531-2538.

60
COSEWIC. 2013. COSEWIC Assessment and Status Report of the Western Rattlesnake
Crotalus oreganus in Canada. Ottawa, ON, Canada: Committee on the Status of
Endangered Wildlife in Canada. Available from: https://www.sararegistry.gc.ca/
virtual_sara/files/cosewic/sr_Western%20Rattlesnake_2015_e.pdf
COSEWIC. 2015. COSEWIC Assessment and Status Report of the Great Basin Gophersnake
Pituophis catenifer deserticola in Canada. Ottawa, ON, Canada: Committee on the
Status of Endangered Wildlife in Canada. Available from: https://www.registrelepsararegistry.gc.ca/virtual_sara/files/cosewic/sr_couleuvre_nez_mince_grd_bassi_gt_b
asin_gophersnake_1213_e.pdf
Cunnington GM, Garrah E, Eberhardt E, Fahrig L. 2014. Culverts alone do not reduce road
mortality in anurans. Ecoscience. 21(2): 69-78.
Crosby JE. 2014. Amphibian occurrence on South Okanagan roadways: Investigating
movement patterns, crossing hotspots, and roadkill mitigation structure use at the
landscape scale. University of Waterloo Masters Thesis. Unpublished.
Donaldson BM, Elliott KEM. 2021. Enhancing existing isolated underpasses with fencing
reduces wildlife crashes and connects habitat. Human–Wildlife Interactions. 15(1):
148–161
Fahrig L, Rytwinski T. 2009. Effects of roads on animal abundance: An empirical review and
synthesis. Ecology and Society. 14(1): 21.
Fernández-López J, Blanco-Aguiar JA, Vicente J, Acevedo P. 2022. Can we model
distribution of population abundance from wildlife-vehicles collision data?
Ecography. 2022:e06113.
Fleet RR, Rudolph DC, Camper JD, Niederhofer J. 2009. Ecological parameters of the
Coluber constrictor etheridgei, with comparisons to other Coluber constrictor
subspecies. Southeastern Naturalist. 8(Special Issue 2): 31-40.
Ford AT, Clevenger AP, Bennett A. 2009. Comparison of methods of monitoring wildlife
crossing-sturctures on highways. Journal of Wildlife Management. 73(7): 1213-1222.
Ford AT, Barrueto M, Clevenger AP. 2017. Road mitigation is a demographic filter for
Grizzly Bears. Wildlife Society Bulletin. 41(4): 712-719.
Forman RT, Sperling TD, Bissonette JA, Clevenger AP, Cutshall CD, Dale VH, Fahrig L,
France R, Goldman CR, Heanue K, et al. 2003. Road Ecology: Science and Solutions.
Washington (DC): Island Press. 481 p.
Gilhooly PS, Nielsen SE, Whittington J, St. Clair CC. 2019. Wildlife mortality on roads and
railways following highway mitigation. Ecosphere. 10(2): e02597.

61
Glista DJ, DeVault TL, DeWoody JA. 2009. A review of mitigation measures for reducing
wildlife mortality on roadways. Landscape and Urban Planning. 91: 1-7.
Harvey JA, Larsen KW. 2020. Rattlesnake migrations and the implications of thermal
landscapes. Movement Ecology. (2020) 8:21.
Jarvis LE, Hartup M, Petrovan SO. 2019. Road mitigation using tunnels and fences promotes
site connectivity and population expansion for a protected amphibian. European
Journal of Wildlife Research. 65: 27.
Lee TS, Creech TG, Martinson A, Nielsen SE, Jakes AF, Jones PF, Sanderson K, Ford AT.
2020. Prioritizing human safety and multispecies connectivity across a regional road
network. Conservation Science and Practice. 2020: e327.
Macartney JM. 1985. The ecology of the Northern Pacific Rattlesnake, Crotalus viridis
oreganus, in British Columbia [thesis]. Victoria, BC, Canada: University of Victoria.
Maida JR, Bishop CA, Larsen KW. 2020. Migration and disturbance: Impact of fencing and
development on Western Rattlesnake (Crotalus oreganus) spring movements in
British Columbia. Canadian Journal of Zoology. 98: 1-12.
Markle CE, Gillingwater SD, Levick R, Chow-Fraser P. 2017. The true cost of partial
fencing: Evaluating strategies to reduce reptile road mortality. Wildlife Society
Bulletin. 41(2): 342-350.
Matson M. 2021. Seasonal and temporal difference in road underpass use for at-risk snakes
in the southern Okanagan [unpublished Honours thesis, Department of Natural
Resource Science]. Kamloops, BC, Canada: Thompson Rivers University.
McGregor RL, Bender DJ, Fahrig L. 2007. Do small mammals avoid roads because of the
traffic? Journal of Applied Ecology. 45(1): 117-123.
McKelvey V. In prep. Natural and artificial hibernacula habitat use of three sympatric snake
species [MSc thesis]. Kamloops, BC, Canada: Thompson Rivers University.
Molyneux J, Pavey CR, James AI, Carthew SM. 2017. The efficacy of monitoring techniques
for detecting small mammals and reptiles in arid environments. Wildlife Research.
44(6-7): 534-545.
Muellman PJ, Cunha OD, Montgomery CE. 2018. Crotalus horridus (Timber Rattlesnake)
maternal scent trailing by neonates. Northeastern Naturalist. 25(1): 50-55.
Pagnucco KS, Paszkowski CA, Scrimgeour GJ. 2011. Using cameras to monitor tunnel use
by Long-toed Salamanders (Ambystoma macrodactylum): An informative costefficient technique. Herpetological Conservation and Biology. 6(2): 277-286.

62
Polak T, Nicholson E, Grillo C, Bennett JR, Possingham HP. 2019. Optimal planning to
mitigate the impacts of roads on multiple species. Journal of Applied Ecology. 56:
201–213.
Plante J, Jaeger JAG, Desrochers A. 2019. How do landscape context and fences influence
roadkill locations of small and medium-sized mammals? Journal of Environmental
Management. 235(2019): 511-520.
Racer Management Team Working Group. 2013. Management plan for the Racer (Coluber
constrictor) in British Columbia. Prepared for the B.C. Ministry of Environment,
Victoria, BC. 25 pp. Available from: https://a100.gov.bc.ca/pub/eirs/viewDocument
Detail.do?fromStatic=true&repository=BDP&documentId=11572
Rouse JD. 2006. Spatial Ecology of Sistrurus catenatus catenatus and Heterodon platirhinos
in a rock-barren landscape. University of Guelph MSc Thesis. Unpublished.
Rytwinski T, Soanes K, Jaeger JAG, Fahrig L, Findlay CS, Houlahan J, van der Ree R, van
der Grift EA. 2016. How effective is road mitigation at reducing road-kill? A metaanalysis. PLoS ONE. 11(11).
Santini L, Saura S, Rondinini C. 2016. A composite network approach for assessing multispecies connectivity: An application to road defragmentation prioritisation. PLoS
ONE 11(10): e0164794.
Sawyer H, Rodgers PA, Hart T. 2016. Pronghorn and Mule Deer use of underpasses and
overpasses along U.S. Highway 191. Wildlife Society Bulletin. 40(1): 211-216.
Seidler RG, Green DS, Beckman JP. 2018. Highways, crossing structures and risk: Behaviors
of Greater Yellowstone pronghorn elucidate efficacy of road mitigation. Global
Ecology and Conservation. 15 (2018).
Smith D, King R, Allen, BJ. 2020. Impacts of exclusion fencing on target and non-target
fauna: A global review. Biological Reviews. (2020)95: 1590-1606.
Southern Interior Reptile and Amphibian Working Group. 2016a. Recovery plan for the
Gopher Snake, deserticola subspecies (Pituophis catenifer deserticola) in British
Columbia. Prepared for the B.C. Ministry of Environment, Victoria, BC. 36 pp.
Available from: https://a100.gov.bc.ca/pub/eirs/viewDocumentDetail.do?fromStatic
=true&repository=BDP&documentId=12590
Southern Interior Reptile and Amphibian Working Group. 2016b. Recovery plan for the
Western Rattlesnake (Crotalus oreganus) in British Columbia. Prepared for the B.C.
Ministry of Environment, Victoria, BC. 37 pp. Available from: https://a100.gov.bc.ca
/pub/eirs/viewDocumentDetail.do?fromStatic=true&repository=BDP&documentId=1
2591

63
Species At Risk Act. 2002. Ottawa, Canada: Minister of Justice; [updated 2023 Feb 3;
accessed 2023 Apr 11]. Available from: https://laws.justice.gc.ca/PDF/S-15.3.pdf
Sunga J, Sayers J, Cottenie K, Kyle CJ, Ethier DM. 2017. The effects of roads on habitat
selection and movement patterns of the American badger subspecies Taxidea taxus
jacksoni in Ontario, Canada. Canadian Journal of Zoology. 95: 821-828.
Theodoratus DH, Chiszar D. 2000. Habitat selection and prey odor in the foraging behavior
of Western Rattlesnakes (Crotalus viridis). Behaviour. 137(1): 119-135.
van der Grift EA, van der Ree R, Fahrig L, Findlay S, Houlahan J, Jaeger JAG, Klar N,
Madriñan, Olson L. 2013. Evaluating the effectiveness of road mitigation measures.
Biodiversity Conservation. 22: 425-448.
Warren SD, Scifres CJ, Teel PD. 1987. Response of grassland arthropods to burning: A
review. Agriculture, Ecosystems & Environment. 19:105-130.
Wearn OR, Glover-Kapfer P. 2019. Snap happy: Camera traps are an effective sampling tool
when compared with alternative methods. Royal Society Open Science. 6: 181748.
Welbourne DJ, Paull DJ, Claridge AW, Ford F. 2017. A frontier in the use of camera traps:
Surveying terrestrial squamate assemblages. Remote Sensing in Ecology and
Conservation. 3(3): 133-145.
Whelan RJ, Rodgerson L, Dickman CR, Sutherland EF. 2002. Critical life cycles of plants
and animals: Developing a process-based understanding of population changes in
fire-prone landscapes. In: Bradstock RA, Williams JE, Gill MA, editors. Flammable
Australia: The Fire Regimes and Biodiversity of a Continent. Cambridge (UK):
Cambridge University Press. p. 94-104.
Winton SA. 2017. Road mortality hot spots and movement corridors of at-risk snake species
in the White Lake Basin of British Columbia. Prepared for the BC Ministry of
Environment. Unpublished.
Winton SA, Bishop CA, Larsen KW. 2020. When protected areas are not enough: Lowtraffic roads projected to cause a decline in a northern viper population. Endangered
Species Research. 41:131–139.
Winton SA, Taylor R, Bishop CA, Larsen KW. 2018. Estimating actual versus detected road
mortality rates for a northern viper. Global Ecology and Conservation. 16 (2018).

CHAPTER 4
CONCLUSION
The overarching goal of my thesis was to understand the immediate impacts of road
mortality mitigation actions on snakes. Specifically, I assessed the numerical and behavioural
(functional) responses of three at-risk grassland snake species in the years immediately
following the installation of ecopassages and drift fencing in the South Okanagan Valley of
British Columbia. The three focal species were Western Rattlesnakes (Crotalus oreganus),
Great Basin Gophersnakes (Pituophis catenifer deserticola), and Western Yellow-bellied
Racers (Coluber constrictor mormon), with a primary focus on rattlesnakes due to longstanding conservation concerns and the relative ease of their population monitoring. I
conducted this research by: (a) comparing rattlesnake roadkill and population estimates preand post-mitigation, (b) quantifying and comparing ecopassage use of three snake species,
and (c) assessing the efficacy of drift fences flanking the ecopassages.
The main findings of my thesis were:
•

Rattlesnake roadkill rates have a decreasing trend over the years from 2015 to 2020,
despite traffic rates increasing.

•

There were no obvious population or survivorship trends for Western Rattlesnakes
throughout the course of the study.

•

In general, the post-mitigation years had less rattlesnake road mortality and a larger
population size than the pre-mitigation years.

•

Gophersnakes showed 1.6x and racers 26x the number of appearances of rattlesnakes
in the ecopassages. Gophersnakes completed 2.4x and racers 62x the number of
passages through ecopassages that rattlesnakes did.

•

The timing of ecopassage use throughout the active season lacked consistency among
species, and within species between years.

•

The use of different ecopassages was not consistent across species, and I postulate
this was based on environmental factors.

•

Rattlesnakes entered fenced ends of ecopassages more than unfenced ends only in the
last year of the study.
64

65
•

Measuring the usage of ecopassages through camera data is a complimentary and
valuable tool to augment the monitoring of demographic trends in the target
population to assess their response.

•

Taken together, the three main components of this study (road mortality,
population/survivorship, and camera data) work together to provide a well-rounded
assessment of the of the response of snakes to mitigation measures, and the
subsequent conservation implications.

Overall, these findings demonstrate that functional responses to newly-installed
mitigation structures can be detected sooner than numerical responses, especially for animals
with relatively long life spans. This study also highlights the importance of (i) monitoring the
response to mitigation efforts (such as ecopassages) immediately after their introduction, but
also (ii) developing long-term data sets that continue well past the ‘shock phase’. The results
of my study depict stark variation in ecopassage use between species, which may be
important to consider when developing species-specific conservation strategies.

Management Implications
Initial management recommendations for the snakes in the study region were outlined
by Winton (2018) following her earlier work on the road ecology of rattlesnakes. She
suggested that fences paired with underpasses be installed at high priority locations to reduce
roadkill while maintaining habitat connectivity. She also emphasized the importance of
continued monitoring of road mortality and the rattlesnake population in order to detect any
changes that result from these mitigation structures. Fortunately, these recommendations
came to fruition and drove the research presented in this thesis. However, there are some
recommendations by Winton (2018) that have not been implemented, and I believe these bear
repeating. Firstly, a reduced speed limit and traffic signs in the basin (as suggested by Farmer
and Brooks 2012, Valero et al. 2015) would slow traffic, raise awareness about snakes’
presence in the area, and draw attention to the impacts of road mortality. Fortunately, ongoing
research in the area (Wong in prep) is assessing the comparative movement patterns of
rattlesnakes, gophersnakes, and racers to better understand their susceptibility to road

66
mortality and affinity to crossing structures. Conducting population-level studies (Chapter 2)
to assess mitigation efforts are more difficult for species that do not easily lend themselves to
population monitoring (i.e. racers and gophersnakes versus rattlesnakes). Examining
movement patterns provides a metric that allows more direct comparisons between the three
species, however it is not a reliable indicator of the conservation efficacy of the mitigation
structures. Additionally, similar research should be done in other areas of concern throughout
the province.
Based on my own research (particularly that described in Chapter 3), I recommend
expanding the length of the fences to better direct snakes into the ecopassages. Partial
fencing has proven to be ineffective, and compromises the success of connectivity structures
(Baxter-Gilbert et al. 2015, Markle et al. 2017). It also results in an issue known as the
“fence-end effect”, where animals are found on the road around fence ends (Harman et al.
2023, Markle et al. 2017). Unfortunately the slope and/or substate of the areas surrounding
the ecopassages at my study site prevented fences from being built on both sides of the
ecopassages. Although extending fences at this site still would result in partial fencing, there
are initial signs of fence success. Therefore, this is a situation where the Precautionary
Principle should be invoked. I also recommend that fences are built with long-term structural
integrity in mind, because hiring personnel to keep up with long-term maintenance will be
expensive and logistically difficult, and compromised fencing has been shown to be
ineffective (Baxter-Gilbert et al. 2015).
In my work, the extreme discrepancy between ecopassage use across the three species
makes it clear that snakes, despite taxonomic and morphological similarity, should not be
expected to respond in similar fashion to mitigation structures. The type of ecopassages
installed were designed for general snake use (BC Ministry of Environment and Climate
Change Strategy 2020, Ontario Ministry of Natural Resources and Forestry 2016), and
therefore provide an ideal structure to use in this study area and other locations where snake
road mortality is of concern. Although the declining population of rattlesnakes at this site
(Winton et al. 2020) was the impetus for the installation of mitigation structures, I have
shown that multiple snake species now use the ecopassages – some more so than

67
rattlesnakes. These snake-targeted ecopassages also happened to accommodate a variety of
other taxa (Jaccard 2024, see Appendix D), which has obvious advantages.

Future Research Considerations
Future research should not only focus heavily on mark-recapture of racers and
gophersnakes, but attempt to use a different capture method, like traps, to increase evenness
in catchability among the sympatric species. This study was principally framed around
rattlesnake conservation, and as a result population estimates for racers and gophersnakes
were unavailable, given a shortage of mark-recapture data for these two species in my study
area. This is largely due to the fact that our mark-recapture survey efforts were focused on
communal rattlesnake hibernacula, and even when present in rattlesnake hibernacula, it
appears they spend comparatively little time there in spring, making capture and enumeration
very difficult (McKelvey in prep). Therefore, it is not possible to determine if the extreme
discrepancy in ecopassage use between species is a result of a corresponding variation in
population size. Since my fieldwork, there have been stronger mark-recapture efforts for
these species in the basin, however the catchability of these snakes (particularly racers)
remains a major obstacle to assessing the status of these animals. Similarly, wildlife camera
detection probablity can vary interspecifically (Caravaggi et al. 2020). Experiments could be
done testing snakes of each species upon ecopassage transit, and identifying the proportion of
times that the cameras are successfully triggered. This would enable the calculation of
correction factors that could be applied to camera capture rates for each species. Finally,
differences in movement and behaviour patterns could be affecting how often each species
encounters ecopassages or fences. As mentioned earlier, a comparison of the habitat use and
movements of rattlesnakes, gophersnakes, and racers is ongoing, in an attempt to relate the
results to the frequency of their ecopassage use (Wong in prep). This research also continues
to monitor rattlesnake roadkill rates and population size, which will add substantially to the
work in this thesis.

68
Conclusion
My research considered both numerical and behavioural (functional) factors to assess
the immediate response of Western Rattlesnakes to ecopassages and drift fencing, along with
the behavioural response of Great Basin Gophersnakes and Western Yellow-bellied Racers.
Although the short timeframe of the ‘shock phase’ following ecopassage installation prevents
my ability to draw strong conclusions from the data, the rattlesnake population estimates
gathered at this time still provide an excellent basis for continued monitoring to generate
longer-term post-mitigation results. Nevertheless, the post-mitigation rattlesnake roadkill
rates are promising, as road mortality appears to be trending downwards despite an increase
in traffic. The behavioural response of the snakes is desirable, as ecopassage use was
immediate and appears to be increasing. Therefore, it is possible that we could see a clear and
positive numerical response in the coming years.
The understanding of the numerical and behavioural responses of a species to
alterations to their environment, both in the short- and long-term, will aid conservation
practitioners when developing recovery strategies for wildlife populations. The results
highlight the importance of long-term monitoring pre- and post- mitigation, and the extreme
variation in responses that can occur across species of the same taxa. I achieved my research
objectives of comparing Western Rattlesnake pre-mitigation roadkill and population
estimates to those in the years immediately following the installation of ecopassages and
fencing, and quantifying and comparing the use of ecopassages by Western Rattlesnakes,
Great Basin Gophersnakes, and Western Yellow-bellied Racers. Based on my results, I make
the following recommendations:
•

Where funding permits, the Precautionary Principle should be applied and
ecopassages and drift fencing should be installed to reduce roadkill of snakes
and other small taxa.

•

Since this site now has the unique and valuable history of intensive
monitoring data pre- and post-mitigation, it should remain a focal research site
for snakes in the Southern Interior of BC. Monitoring of snake roadkill and
rattlesnake population size should be continued long-term.

69
•

Analyzing camera data is labour intensive, and given my study shows quite
clearly that snakes are using the ecopassages, the value of ongoing camera
monitoring is dubious, particularly if it deflects money and energy away from
other investigations. Software for analyzing wildlife photos is currently not
effective for distinguishing snakes (K.W. Larsen, pers. obs.). For these
reasons, I would recommend ceasing camera monitoring in the ecopassages in
order to redirect resources to other areas of the project, or at least move away
from a yearly monitoring scheme.

•

Monitor Great Basin Gophersnake and Western Yellow-bellied Racer
populations in a way that can be compared among species, accounting for
differences in catchability.

•

Monitor movement patterns of all three species to gain insight into the
differences in their ecopassage use.

In closing, I have met the objectives of my thesis, and I trust that the findings of my research
will contribute to the conservation and persistence of the snake populations in the White
Lake Basin, and to wildlife conservation in general.

70

LITERATURE CITED
Baxter-Gilbert JH, Riley JL, Lesbarrères D, Litzgus JD. 2015. Mitigating reptile road
mortality: fence failures compromise ecopassage effectiveness. PLoS ONE 10(3).
BC Ministry of Environment and Climate Change Strategy. 2020. Guidelines for amphibian
and reptile conservation during road building and maintenance activities in British
Columbia. Version 1.0. 151p.
Caravaggi A, Burton AC, Clark DA, Fisher JT, Grass A, Green S, Hobaiter C, Hofmeester
TR, Kalan AK, Rabaiotti D, et al. 2020. A review of factors to consider when using
camera traps to study animal behavior to inform wildlife ecology and conservation.
Conservation Science and Practice. 2(8): e239.
Farmer RG, Brooks RJ. 2012. Integrated risk factors for vertebrate roadkill in southern
Ontario. The Journal of Wildlife Management. 76:1215–1224.
Fitch HS. 1987. Collecting and life-history techniques. In: Seigel RA, Collins JT, and Novak
SS, editors. Snakes: Ecology and Evolutionary Biology. New York: Macmillan. p.
143–164.
French SS, Webb AC, Hudson SB, Virgin EE. 2018. Town and country reptiles: a review of
reptilian responses to urbanization. Integrative and Comparative Biology. 58(5): 948966.
Harman KE, Omlor K, Popescu V. 2023. Evaluating fence-end treatments for migratory
amphibians. Frontiers in Ecology and Evolution. 11: 1008462.
Jaccard S. 2024. Analyzing the use of snake eco-passages by non-target species in White
Lake Basin, British Columbia [unpublished directed study, Department of Natural
Resource Science]. Kamloops, BC, Canada: Thompson Rivers University.
Jeminson SC, Bishop LA. May PG, Farrell TM. 1995. The impact of PIT-tags on growth and
movement of the rattlesnake, Sistrurus miliarius. Journal of Herpetology. 29(1): 129132.
Krebs CJ. 1989. Ecological Methodology. New York (NY): HarperCollinsPublishers, Inc.
654 p.
Markle CE, Gillingwater SD, Levick R, Chow-Fraser P. 2017. The true cost of partial
fencing: evaluating strategies to reduce reptile road mortality. Wildlife Society
Bulletin. 41(2): 342-350.
McKelvey V. In prep. Natural and artificial hibernacula habitat use by three sympatric snake
species [MSc thesis]. Kamloops, BC, Canada: Thompson Rivers University.

71
Ontario Ministry of Natural Resources and Forestry. 2016. Best management practices for
mitigating the effects of roads on amphibians and reptile species at risk in Ontario.
Queen’s Printer for Ontario. 112 p.
Valero E, Picos J, Lagos L, Álvarez X. 2015. Road and traffic factors correlated to wildlife–
vehicle collisions in Galicia (Spain). Wildlife Research. 42:25-34.
Winton S. 2018. Impacts of road mortality on the Western Rattlesnake (Crotalus oreganus) in
British Columbia. Thompson Rivers University MSc Thesis. Unpublished.
Wong C. In prep. Movement ecology and road mortality for a community of threatened
snakes in British Columbia, Canada [MSc thesis]. Kamloops, BC, Canada: Thompson
Rivers University.

APPENDIX A
JOLLY-SEBER ESTIMATES FOR RATTLESNAKES IN THE WHITE LAKE BASIN
Table A.1. Jolly-Seber population estimates for Western Rattlesnakes (Crotalus oreganus) in
the White Lake Basin, BC (2015-2020; estimates not available for 2015 or 2021 based on
method of calculation).

All Captures

All age
classes

Adults only

Focal Dens

Year

Population
Size

SE

Population
Size

SE

2015

NA

NA

NA

NA

2016

644

70

526

54

2017

730

86

497

53

2018

1038

128

876

107

2019

1365

152

900

96

2020

834

90

666

77

2021

NA

NA

NA

NA

MEAN

922

105

693

77

2015

NA

NA

NA

NA

2016

382

39

325

31

2017

393

41

272

25

2018

475

55

335

35

2019

611

72

405

47

2020

386

40

283

31

2021

NA

NA

NA

NA

MEAN

449

49

324

34

72

Table A.2. Jolly-Seber survivorship estimates for Western Rattlesnakes (Crotalus oreganus)
in the White Lake Basin, BC (2015–2021; estimates not available for 2020 or 2021 based on
method of calculation).

All Captures

All age
classes

Adults only

Focal Dens

Year

Survival
Estimate

SE

Survival
Estimate

SE

2015 → 2016

0.72

0.07

0.71

0.06

2016 → 2017

0.64

0.07

0.62

0.06

2017 → 2018

0.73

0.08

0.83

0.09

2018 → 2019

0.75

0.08

0.71

0.07

2019 → 2020

NA

NA

NA

NA

2020 → 2021

NA

NA

NA

NA

MEAN

0.71

0.08

0.72

0.07

2015 → 2016

0.72

0.06

0.71

0.06

2016 → 2017

0.66

0.06

0.65

0.06

2017 → 2018

0.73

0.07

0.81

0.08

2018 → 2019

0.93

0.10

0.95

0.10

2019 → 2020

NA

NA

NA

NA

2020 → 2021

NA

NA

NA

NA

MEAN

0.76

0.07

0.78

0.08

73

APPENDIX B
SNAKE ROAD MORTALITY IN THE WHITE LAKE BASIN

Table B.1. Frequency of roadkill detected for at-risk snake species during surveys and
incidentally along 11.7 kms of road within the White Lake Basin, BC, Canada.

Year
2015
2016
2017
2018
2019
2020

CROR
36
56
28
29
35
21

PICA
32
52
47
60
56
56

COCO
49
79
77
87
38
47

74

Table B.2. Detections of dead Western Rattlesnakes (Crotalus oreganus) and densities per km surveyed (unadjusted for observer error
or scavenging rates) by road survey method in the White Lake Basin, BC, Canada, 2015-2018.

Survey
method

Time
of day

Total km surveyed

Number of dead rattlesnakes detected

Unadjusted dead
rattlesnake density
(dead rattlesnakes/km)

2015

2016

2017

2018

2019

2020

2015

2016

2017

2018

2019

2020

2015

2016

2017

2018

2019

2020

Walk

Day

116.2

253.5

485.6

274.8

386.1

538.2

6

13

8

10

20

16

0.052

0.051

0.016

0.036

0.052

0.030

Drive

Day

62.4

51.6

157.9

81.1

210.6

35.1

3

4

2

5

7

1

0.048

0.078

0.013

0.062

0.033

0.028

Drive

Night

233.3

292.5

222.3

81.9

0

0

12

9

2

5

-

-

0.051

0.031

0.009

0.061

-

-

Bike

Day

53.1

129.9

0

64.5

23.4

0

1

8

-

1

1

-

0.019

0.062

-

0.016

0.085

-

Surveys total

465

727.5

865.8

502.3

620.1

573.3

22

34

12

21

28

17

0.047

0.049

0.014

0.042

0.045

0.030

Incidental

1633.8

2382.4

-

-

-

-

14

22

16

8

7

4

0.009

0.009

-

-

-

-

Surveys +
Incidental

2098.8

3109.9

-

-

-

-

36

56

28

29

35

21

0.017

0.018

-

-

-

-

75

Table B.3. Calculated road mortality rates (deaths/km/day) and number of Western
Rattlesnake (Crotalus oreganus) deaths based on walking survey results and accounting for
scavenger-removal and observer error during the active season (April-October) along the
11.7 km survey route in the White Lake Basin, BC, Canada, 2015-2018.

After
culvert
installation

After
fencing
installation

2015

2016

2017

2018

2019

2020

0.044
(0.019)

0.070
(0.030)

0.043
(0.019)

0.021
(0.006)

0.039
(0.010)

0.025
(0.007)

Active season length
(days)

176

188

183

182 *

182

181

Calculated rattlesnake
deaths per year (11.7
km survey route)

91

154

93

45

83

53

Number of dead
rattlesnakes detected

36

56

28

29

35

21

Correction factor

2.5

2.8

3.3

1.6

2.4

2.5

Mean mortality rate
(±SE) (rattlesnake
deaths/km/day)

*Average of 2015-2017 due to late spring sampling in 2018

76

Table B.4. Adjusted dead Western Rattlesnake (Crotalus oreganus) road densities (dead rattlesnakes/km) for five types of survey
methods, accounting for scavenger-removal and observer error, in the White Lake Basin, BC, Canada, 2015-2018.

Survey method

Time of day

Walk

Adjusted dead rattlesnake density (dead rattlesnakes/km)
2015

2016

2017

2018

2019

2020

Day

0.130

0.143

0.054

0.058

0.125

0.075

Drive

Day

0.120

0.218

0.044

0.099

0.079

0.070

Drive

Night

0.128

0.087

0.031

0.098

-

-

Bike

Day

0.048

0.174

-

0.031

0.204

-

Surveys total

0.118

0.137

0.046

0.068

0.108

0.075

Incidental

0.023

0.025

-

-

0.007

0.005

Surveys + Incidental

0.043

0.050

-

-

0.026

0.018

77

APPENDIX C
SNAKE ECOPASSAGE USE IN THE WHITE LAKE BASIN
Table C.1. The distribution of ecopassage use of Western Yellow-bellied Racers, Great Basin Gophersnakes, and Western
Rattlesnakes in the White Lake Basin, British Columbia, Canada, with the proportion of the species total in brackets. An appearance
was defined as any camera detection of the species inside the ecopassage and a passage occurred when a snake of the same species
entered and exited opposite ends of the ecopassage within 30 minutes.
Racer
Year

2019

2020

Ecopassage
1
2
3
4
5
6
7
8
1
2
3
4
5
6
7
8

Appearance
4 (0.003)
145 (0.119)
22 (0.018)
335 (0.275)
203 (0.167)
377 (0.309)
60 (0.049)
73 (0.060)
96 (0.052)
126 (0.068)
137 (0.074)
588 (0.319)
346 (0.188)
336 (0.182)
101 (0.055)
112 (0.061)

Passage
1 (0.003)
44 (0.127)
1 (0.003)
161 (0.465)
2 (0.006)
124 (0.358)
6 (0.017)
7 (0.020)
4 (0.010)
34 (0.085)
19 (0.047)
157 (0.391)
137 (0.341)
37 (0.092)
1 (0.002)
13 (0.032)

Gophersnake
Appearance
Passage
0 (0)
0 (0)
3 (0.041)
0 (0)
16 (0.216)
0 (0)
8 (0.108)
3 (0.428)
15 (0.203)
2 (0.286)
13 (0.176)
2 (0.286)
4 (0.054)
0 (0)
15 (0.203)
0 (0)
12 (0.110)
0 (0)
4 (0.037)
2 (0.091)
22 (0.202)
2 (0.091)
13 (0.119)
5 (0.227)
21 (0.192)
6 (0.273)
5 (0.046)
1 (0.045)
4 (0.037)
0 (0)
28 (0.257)
6 (0.273)

Rattlesnake
Appearance
Passage
19 (0.358)
0 (0)
6 (0.113)
2 (0.667)
1 (0.019)
0 (0)
0 (0)
0 (0)
15 (0.283)
0 (0)
6 (0.113)
0 (0)
4 (0.075)
1 (0.333)
2 (0.038)
0 (0)
22 (0.355)
2 (0.222)
11 (0.177)
2 (0.222)
6 (0.097)
0 (0)
0 (0)
0 (0)
14 (0.226)
3 (0.333)
6 (0.097)
2 (0.222)
3 (0.048)
0 (0)
0 (0)
0 (0)

78

APPENDIX D
ECOPASSAGE USE BY NON-TARGET SPECIES IN THE WHITE LAKE BASIN

Table D.1. Ecopassage camera appearances and passages of non-target species in 8
ecopassages in the White Lake Basin, British Columbia, Canada in 2019 and 2020. (Adopted
from Jaccard 2024)
Appearances

Passages

2019

2020

2019

2020

American Badgers

1

0

0

0

Birds

22

99

0

0

Burrowing Owls

1

0

0

0

Bushy-tailed Woodrats

33

1

2

0

Frogs & Toads

22

12

0

0

Muskrats

2

0

0

0

Porcupines

2

0

0

0

Rabbits

72

10

0

0

Salamanders

29

12

4

0

Skunks

3

0

2

0

Weasels

17

29

4

16

Western Painted Turtles

3

0

2

0

Yellow-bellied Marmots

0

6

0

2

Yellow-pine Chipmunks

7

1

0

0

>200

>200

-

-

Mice/ Voles

79