-
Title
-
Roads and raptors: A spatial analysis of reported raptor collisions on the Washington state highway system
-
Date
-
2021
-
Creator
-
Croston, Sarah
-
Identifier
-
Thesis_MES_2021_CrostonS
-
extracted text
-
ROADS AND RAPTORS
A SPATIAL ANALYSIS OF REPORTED RAPTOR COLLISIONS ON THE WASHINGTON
STATE HIGHWAY SYSTEM
by
Sarah L. Croston
A Thesis
Submitted in partial fulfillment
of the requirements for the degree
Master of Environmental Studies
The Evergreen State College
June 2021
©2021 by Sarah L. Croston. All rights reserved.
This Thesis for the Master of Environmental Studies Degree
by
Sarah L. Croston
has been approved for
The Evergreen State College
by
John C. Withey, Ph.D.
Member of the Faculty
June 7, 2021
Date
ABSTRACT
Roads and Raptors:
A Spatial Analysis of Reported Raptor Collisions on the Washington State Highway System
Sarah L. Croston
Human activity impacts animals in many negative ways. As our infrastructure demands grow,
habitat loss and fragmentation are inevitable. When available habitat shrinks and ecosystems
become less connected, animals are faced with complex challenges. One of the single most
devastating anthropogenic disturbances in modern times has been the expansion of road
networks. Vehicle strikes kill millions of animals each year worldwide, and roads also act as a
barrier for flying species such as bats and birds. Birds experience high rates of mortality due to
vehicle collisions. Raptors (birds of prey) are highly susceptible to vehicle collisions due to their
hunting and feeding behaviors. My research examined locations of raptors hit on the Washington
State Highway System from 2015-2020. Using the Washington Department of Transportation’s
Wildlife Carcass Removal Database, I created a series of maps in ArcGIS Pro, and used a kernel
density function to identify hotspots on major highways in the state. Four hotspot locations were
found for reported raptor collisions. Two of these locations were on Interstate 5, and the other
two were found on Interstate 90. My research led me to explore the realm of community science
efforts in roadkill data due to some limitations I experienced with the wildlife carcass removal
database. I created a survey in ArcGIS’s Survey 123 platform, making a data collection tool to
expand efforts around raptor collisions in the state. My spatial analysis results may help
strengthen information and spread awareness of raptor vehicle collisions in the future. This could
lead to mitigation efforts in specific spots on Washington State Highways where raptors are
being killed in large quantities due to vehicle collisions.
Table of Contents
List of Figures………………………………………………….…………………………vi
List of Tables………………………………...………………………………………….viii
Acknowledgments……………………………………..…………………………………ix
Introduction………………………………………………………………………..………2
Literature Review……………………………………………………………………..…...6
Introduction…………………………………………………………………..……6
Raptor Ecology……………………………………………………………………7
Habitat Loss and Fragmentation…………………………………..…………….11
Animal Collisions………………………………………………………….……..13
Mitigation Efforts………………………………………………………………...20
Methods…………………………………………………………………………………..26
Study Area & Data Sources ……………………………………………………..26
ArcGIS Pro Applications………………………………………………………...31
Kernel Density Analysis………………………………………………………….31
ArcSurvey 123 Application………………………………………………………32
Limitations……………………………………………………………………….32
Results……………………………………………………………………………………34
Discussion…………………………………………………………………………..……45
Hotspots in Washington …………………………………………………………48
Gaps in Current Research……………………………………………………….49
Future Studies……………………………………………………………………49
Conclusion……………………………………………………………………………….54
Bibliography……………………………………………………………………………..56
Appendices………………………………………………………………………………67
Appendix A: Washington State Highway Amount of Vehicle Miles Traveled per year
1990-2016………………………………………………………………......67
iv
Appendix B: Complete list of Raptor Species included in Arc Survey123……….68
v
List of Figures
Figure 1. Status of the world’s raptor populations. Above, by Red List Categories LC-Least Concern, NTNear Threatened, VU-Vulnerable, EN-Endangered, CR-Critically Endangered, DD-Data Deficient;
below, by population trend. Figure 1 in McClure et al. 2018………..……………………………….........9
Figure 2. Threats to raptors by group. Figure 6 in McClure et al. 2018…………….…………….……..10
Figure 3. The spatial separation of habitats becoming less connected through the process of habitat
fragmentation. Figure 1 in Fahrig 2003………………….…………………….…………………….…..12
Figure 4. An example of a temporal dynamic sign that is used in the program, Give Wildlife a Brake.
Retrieved from The Jackson Hole Wildlife Foundation’s website………………………………….……..22
Figure 5. A map of the study area, Washington State Highways shown in pink. Created in ArcGIS
Pro………………….…………………………………………………………………...…...…...…....…..27
Figure 6. An example of the roadkill spreadsheets before the data was collected by iPad. Retrieved from
WSDOT………………….……………………………………………...……………………….…….…..27
Figure 7. A map displaying the different regions of Washington, a field that is collected in the wildlife
carcass removal database. Retrieved from WSDOT…………………………..............................………..29
Figure 8. Owls in Washington state, a key to help identify owl carcasses. Created by Habitat Connectivity
Biologist Glen Kalisz, Retrieved from WSDOT…………………………………………………………...30
Figure 9. The expanded equation to calculate kernel density Retrieved from esri, n.d. ...…………….…31
Figure 10. Raptor types reported in each region of the state in the WCRDB from 2015-2020…………..36
Figure 11. A graphic display of raptor collisions by county reported in the WCRDB from 20152020………………………………………………………………………...……………………….……..37
Figure 12. All reported raptor carcasses displayed spatially from 2015-2020. Map was created in ArcGIS
Pro………………………...………………………………..……………………...……………………....38
Figure 13. A spatial display to document the locations and frequencies of owl records with and without
photos attached in the HATS database. Map was created in ArcGIS Pro………………………...………38
Figure 14. A spatial display to document the locations and frequencies of hawk records with and without
photos attached in the HATS database. Map was created in ArcGIS Pro………………………...………39
Figure 15. Kernel Density distribution of raptor collisions in Washington. Map was created in ArcGIS
Maps………………….…………………………………………………………………………..…....…..40
Figure 16. A close-up view of raptor the I5 raptor hotspot A. Map was created in ArcGIS Pro………...41
vi
Figure 17. A close-up view of raptor the I5 raptor hotspot B. Map was created in ArcGIS Pro……….42
Figure 18. A close-up view of raptor the I90 raptor hotspot A. Map was created in ArcGIS Pro……...43
Figure 19. A close-up view of raptor the I90 raptor hotspot B. Map was created in ArcGIS Pro……….44
Figure 20. The amount of vehicle miles traveled (AVMT) in Washington state from 1990-2016. Data is
separated by rural and urban miles. Graph is taken from 2016 Annual Traffic Report, Washington
Department of Transportation, pp.55. ………………….……………………....................................…..47
Figure 21. A systematic way to select objectives of a WVC survey. Figure 62.1 Shilling et al. 2015.
………………….…………………………………………………………………...………………..…..50
vii
List of Tables
Table 1. Survey methods used in studies of road effects on wildlife. For each method type, there are a
number of factors involved that explore how effective the survey method is. These survey types are varied
and range in scale and effort. It is important to recognize that different methods can be useful depending
on the size and scope of a road ecology study. Retrieved from Smith & van der Ree.
2015……………………………….…………………………………..……………………………...……15
Table 2. Estimate of yearly bird mortality events in the United States. Retrieved from Erickson et al.
2005……..................................................................................................................................................…18
Table 3. A visual representation of different problems roads create and solutions to fit the needs of
particular groups of birds. Retrieved from Jacobson, 2005………………………………………….…...26
Table 4. The total recorded raptor carcasses in the WCRDB from 2015-2020, broken up by whole number
as well as percent…………………………………………………………..…………………………...…34
Table 5. Reported raptor carcasses in the WCRDB from 2015-2020, broken up into raptor group and
separated by year…………………………………………………………………………………………35
Table 6. Reported owl carcasses in the WCRDB from 2015-2020, as monthly
totals.……………………………………………………………………………...……...……………..…35
Table 7. A display of all owl records with species written in the comments section of the WCRDB from
2015-2020, by WSDOT Region (see Figure 7)……………………………...…….……………………....36
viii
Acknowledgments
I would like to extend my deepest gratitude to the Washington Department of Transportation’s
habitat connectivity branch. The experience I have gained through my internship under the
guidance of Glen Kalisz has been instrumental in building my thesis and deepening my
understanding of road ecology. Thank you, Glen, for your support throughout my internship and
thesis process and for instilling a road ecology mindset that has been influential in my master’s
program. This internship has been the highlight of my graduate school experience, and I hope to
carry this knowledge into my professional career.
Thank you to my fiancé for being my rock during the thesis process. I appreciate your
unwavering patience, support, and for allowing me to turn our kitchen table into a library of
everything road ecology for the past year.
I would like to thank Mike Ruth for the proficiencies and tools I learned throughout my GIS
classes at Evergreen. Because of these classes, I felt empowered in my GIS abilities and could
use them throughout my thesis process; without the GIS skillset, I would not have been able to
articulate my finding in such a powerful way. I am grateful for your classes and seeing how data
can be transformed into a story through map creation.
Lastly, I want to extend my sincere appreciation to my thesis reader John Withey; thank you for
your help, feedback, and support before and throughout my thesis process. It was in your Urban
Ecology class I began thinking about the relationship between birds and human infrastructure,
which helped to frame and guide my thesis work into fruition.
ix
Introduction
“Roads appear as major conspicuous objects in aerial views and photographs, and their
ecological effects spread through the landscape. Few environmental scientists, from population
ecologists to stream or landscape ecologists, recognize the sleeping giant, road ecology. This
major frontier and its applications to planning, conservation, management, design, and policy are
great challenges for science and society.”
-Richard T.T Forman & Lauren E. Alexander, Roads and Their Major Ecological Effects (1998)
Clad in orange, glowing like a highlighter, I make my way down the side of the highway.
Pressed up against the guardrail, feeling like I could get blown off my feet as semi-trucks barrel
past me, my heart beats faster. Between the off-putting smell, the excessive amount of litter, and
the distorted cloud of constant noise, all I want is an escape from this environment. This is no
place for animals, crossing the highway like a game of chicken. No bright orange color draped
around their bodies, no crosswalks, no bright flashing lights, no crossing guards to help them get
across safely.
Humans have carved up the land, laying down ribbons of roadways, disconnecting the
natural landscape, pushing wildlife into smaller and smaller patches of intact habitat. The need to
study, understand, and mitigate risks associated with roads reaches all corners of the world.
Currently, 750 million vehicles operate on these roads, and their numbers are increasing (van der
Ree et al 2011). This linear infrastructure disconnects habitat for wildlife and increasing the
interface between animals and roads. Roads pose enormous risks for animals as road networks
span over 31 million miles connecting human environments across the globe (van der Ree et al
2011).
1
Even though humans understand that roads are and historically have been a significant
impediment to animals, the branch of road ecology science is relatively new and small. The term
“road ecology” was first used in 1981; it was translated from German to English for the seminal
book, Road Ecology: Science and Solutions, published in 2003 by Richard T.T. Forman and
coauthors (van der Ree et al. 2011). Transportation planners must understand numerous
variables associated with roads, and as outlined in Forman et al. (2003, pp. 99), the objectives are
to:
•
Minimize cost
•
Maximize motorist safety
•
Enhance visual quality while maintaining ecological benefits
•
Reduce erosion and sediment flow
•
Control non-native species
•
Enhance biodiversity
•
Enhance wildlife density and reduce the road barrier effect to improve animal crossing of
roads
•
Reduce wildlife density and increases the road barrier effect to reduce roadkills and
wildlife-vehicle crashes
•
Accomplish a multiple-use array of societal goals
This list consists of highly lofty objectives which cover a lot of ground and reflect the
complicated nature of the roads. As Forman et al. (2003) state and van Ree restates in 2011,
scientists and transportation planners must work together for the future of roads and wildlife (van
Ree et al. 2011). Many transportation agencies prioritize sustainability as an objective that helps
2
enforce policies to ensure the safety of wildlife and humans near and on roads. In 2007, The
Washington Department of Transportation (WSDOT) created a policy which was issued to
protect habitats and wildlife alike as stated in WSDOT Secretary’s Executive Order 103:
Assume that road and highway programs recognize, together with other needs, the
importance of protecting ecosystem health, the viability of aquatic and terrestrial wildlife
species, and the preservation of biodiversity.
Looking forward, executive order 1031 will be vital in protecting habitat and wildlife in
the state. Washington state is currently experiencing population growth; right now, Washington
ranks thirteen in population numbers in the United States. Washington is experiencing a growth
rate of 1.27%, ranking eighth in the US (World Population Review). This is leading to increased
development and habitat loss and fragmentation are inevitable. With more than 50 percent of the
land is owned by private companies and individuals, developers need to understand which
habitats are being built upon and where expansion occurs (Washington Department of Fish and
Wildlife, 2009). Each region that gets developed supports many species as Washington state is
home to numerous different ecoregions.
My research aimed to gather baseline data to understand better where raptors are being
killed on Washington State Highways. I wanted to identify raptor vehicle collision hotspot
locations. The questions I sought out to answer in this work include:
1. Where are raptors being hit and killed due to vehicle collisions on Washington State
Highways?
2. What are the emerging temporal and spatial trends in the data? Are there hotspots that
appear? Are there certain times of year in which more raptor collisions are reported?
3
3. What is the methodology in capturing raptor data within the wildlife carcasses removal
database? Are there gaps in the data? How can this data source be strengthened?
To answer these questions, I created a series of maps in both ArcGIS Pro and ArcGIS
Maps. I also used a kernel density function to identify hotspots based on collision locations.
Additionally, I looked to create a way to collect informed data across the state using the ArcGIS
Survey 123 Platform. I analyzed how a community science effort could further strengthen the
records of raptors collected.
This thesis consists of five sections. The first is a literature review to situate my research
in the broader context of road ecology and provide background information. The literature
review aims to connect the realms of raptor ecology, habitat loss, and fragmentation, wildlifevehicle collisions, mitigation efforts surrounding animal collisions, and educate the reader on
current gaps in this area of research. The methods section explains the process of gathering,
cleaning, and exporting the raptor collision data from 2015-2020. This section also details map
creation and hotspot analysis in Arc GIS Pro and Arc GIS Maps. The results and discussion
sections explore the findings and patterns in the data. The results section includes an array of
tables, graphs, and maps to display where raptors are being hit on the Washington state highway
system. The discussion section elaborates on trends and hotspots found in the spatial analysis.
This section also delves into gaps in the current research of road ecology and raptor road
relations. Lastly, the discussion section looks to the future and what work could be done to
support more findings in the interface of birds and collisions utilizing community science data
collection methods. The final section of the paper concludes the findings in the results of this
work. This field of work lends itself to more significant decisions that will need to be looked at
in the future by transportation planners and ecologists alike.
4
Literature Review
Introduction
Humans are ecosystem engineers. We alter the landscape to best suit our needs; as human
beings redesign the environment; wildlife needs are pushed to the side. People have addressed
the desire for connectivity by making roads. In the United States, nearly 20% of the land is
directly affected by public roads (Husby 2016). The infrastructure humans have created lessened
intact habitats for animals. More roads mean less space for wildlife. The interface of edge habitat
and roads has led to many vehicle animal collisions.
Animal collisions are not only dangerous to wildlife but to humans as well. Copious
numbers of animals are struck and killed by motor vehicles each year. As more roads are
constructed, there is an increase in the number of drivers on the roads. In California alone, an
estimated 8.4 large animals are killed per day, and State Farm Insurance has reported upwards of
23,000 claims per year for accidents involving deer in that state (Nguyen et al. 2020).
The presence of roads affects animal behavior near and far. The visible effects of roads
such as animal carcasses only tell a portion of the story. Animals changing their behavior to
avoid roads account for more of the total disturbance that roads create to animal movements and
behavior (Forman & Alexander 1998; Hovick et al. 2014).
Humans have tried to negate these animal collisions in hotspot areas (areas where many
animals are hit). Wildlife crossing structures have been constructed and are being monitored for
wildlife use. Wildlife crossing structures do not specifically address the needs of flying animals,
5
birds in particular. In the United States alone, it is estimated that between 80-340 million birds
are involved in car collisions each year (Loss et al. 2014).
This literature review will examine raptor ecology and wildlife-vehicle collisions with a
focus on avian species. It will also address the role of habitat fragmentation and habitat loss.
Lastly, it will examine road ecology mitigation efforts.
Raptor Ecology
Raptors (birds of prey) hold an important niche within their environments (Burfield
2008). Occupying a high trophic level as top predators and scavengers, raptors provide a suite of
ecosystem services in their environments (Donázar et al. 2016; HawkWatch International;
Meunier et al. 2000; McClure et al. 2018). Raptors control small mammal populations, acting as
a system of checks and balances (Government of Alberta 2002). Vultures will be categorized as
raptors throughout the length of this paper, as most avian resources include vultures as diurnal
raptors (McClure et al. 2019). Raptors that scavenge, such as eagles and vultures, help keep the
ecosystem clean and assist in decomposition.
Birds of prey are also important indicator species, meaning we can tell how well the
ecosystem functions by the fitness of raptors (Donázar et al. 2016; McClure et al. 2018). The
high trophic level that predatory species occupy makes them sensitive to changes in the
ecosystem (Kovacs et al. 2008). By analyzing the patterns within raptor populations, we can
better understand how an ecosystem is faring as a whole. When conservation plans are situated
around raptors, the plans have positive results as raptors are widespread, easy to observe,
susceptible to changes in their surroundings, and popular within society. (Burfield 2008; Donázar
et al. 2016; Kovacs et al. 2008). Conservation policies that start at a very high trophic level with
6
predatory birds can encapsulate the surrounding flora and fauna as well. Animals that occupy
higher trophic levels are not as plentiful in an ecosystem because they consume a plethora of
resources; thus, each individual is essential for a stable population.
Raptors are long-lived, require ample space for their habitat, and are highly affected by
anthropogenic disturbances (Mcclure et al. 2018). One study in Southern California found that
short-eared owls nesting in areas experiencing high development rates had caused a dramatic
decline in their population, at least a 55% loss in numbers of these nocturnal raptors due to
urbanization (Bloom & McCravy 1996). This is just one example that speaks volumes for raptors
failing to adapt in an area of rapid growth.
Raptors can adjust if suitable measures are put into place, even in the case of the
Northern Spotted Owl, which has precise habitat requirements. Northern Spotted Owls are
typically found in old-growth forests; however, when replanted forests consider the needs of
spotted owls, the owls can be successful if the outcome of the newly planted forest provided the
owl with the form and fit that they seek in old-growth environments (Petty 1996; Horton 1996).
Even though raptors can adapt to changes in the environment, there is less space available
for them; worldwide, raptors are in decline, as McClure and colleagues write (2018).
7
Figure 1. Status of the world’s raptor populations. Above, by Red List Categories LC-Least Concern, NTNear Threatened, VU-Vulnerable, EN-Endangered, CR-Critically Endangered, DD-Data Deficient;
below, by population trend. Figure 1 in McClure et al. 2018.
Most owl populations worldwide are either stable or decreasing, with hawk and eagle
populations experiencing a similar trend (Figure 1, McClure et al. 2018). 18% of raptors are
threatened with extinction, and 52% of raptors have declining populations (McClure et al. 2018).
Declining raptor populations are chiefly due to habitat being transformed into agricultural land
and forested lands being logged (Figure 2, McClure et al. 2018). As landscapes change to
agricultural fields and actively logged forests, this creates patches of unusable habitat for raptors.
The land cover changes and become fragmented.
8
Figure 2. Threats to raptors by group. Figure 6 in McClure et al. 2018.
A study in an area that has been converted to agricultural land in France examined the
relationship of raptors uses the sides of roads for hunting purposes. Meunier et al. researched
diurnal birds of prey and their relationship to roadsides; there is a lack of research that looks at
this association. Diurnal raptors were more present at the roadsides in the wintertime than any
other season to hunt small mammals along roads. This study concluded that more research
should be conducted to better understand how raptors use roadsides in highly converted
9
farmlands because there is high prey availability alongside roads and in medians (Meunier et al.
2000).
Habitat Loss and Fragmentation
It is crucial to define habitat loss, and habitat fragmentation, in any comprehensive
treatment of the topic. Roads play a significant role in creating smaller disconnected habitats and
decreasing the size of intact landscapes. Landscapes, in general, are variable; they differ in land
cover, size, and distribution (Collinge 2009: Fahrig 1997). Humans and wildlife view landscapes
and space through entirely different lenses because our interactions with the land are much
different than wildlife and the environment (Lindenmayer & Fischer 2006). This has led to
humans significantly altering the land in ways we see fit, while animals must adapt to these
changes.
Habitat loss and fragmentation are complex principles to separate because many studies
have not been able to discern the effects of habitat loss from those of habitat fragmentation
(Collinge 2009). For the purposes of this research, we will utilize the definitions outlined in
Ecology of Fragmented Landscapes. Author Sharon Collinge defines habitat loss as “anytime a
piece of land is converted from its current state to some other land use or land cover type.”
Whereas habitat fragmentation “denotes a particular spatial process of land conversion.” (pp. 3).
The process of habitat fragmentation is illustrated in (Figure 3, Fahrig 2003); over time, the
intact habitat is broken up into smaller, spatially separate parcels of land.
10
Figure 3. The spatial separation of habitats becoming less connected through the process of habitat
fragmentation. Figure 1 in Fahrig 2003.
Although natural effects can lead to habitat loss and fragmentation, anthropogenic
urbanization and the transformation of landscapes into agricultural fields have a much more
devastating impact on the environment (Collinge 2009: Lindenmayer & Fischer 2006). Habitats
can recover from specific temporary loss and fragmentation in cases of logging and wildlife
fires. It is unlikely that habitats will bounce back from the pressures of urbanized development
and industrial, agricultural use (Lindenmayer & Fischer 2006).
The process of habitat fragmentation has adverse effects on wildlife, including raptors.
As intact areas decrease in size, the richness of species decreases and negatively affects
biological variety (Hinam & Clair 2008; Lindenmayer & Fischer 2006; Tigas et al. 2002;
Wilcove et al. 1986). Fragmented areas may, in some cases, be too small to sustain populations
of species (Tigas et al. 2002). The process of habitat fragmentation can be connected to the
extinction of species over time (Patten et al. 2005). Birds are affected by the changes in habitat
size and shape.
Birds in altered landscapes face several challenges during the breeding season. Species
may experience shorter breeding seasons and lay fewer eggs which are lighter in weight (Hinam
11
& Clair 2008). As it is taxing to move between smaller parcels, birds experience decreased
numbers of chicks that tend to be smaller; this is influenced by the food limitations that come
with less available space (Hinam & Clair 2008: Lindenmayer & Fischer 2006). These changes
add up quickly and can substantially impact the livelihood of birds in fragmented habitats well
into the future. The breeding season effects of habitat loss and fragmentation are just one season
in the lives of wildlife. Animals also require different size parcels of land for dispersing,
foraging, and migrating (Collinge 2009: Lindenmayer & Fischer 2006). An area that requires
further research is understanding how individuals are affected by habitat fragmentation, as
habitat loss and fragmentation are usually studied at the population level (Hinam & Clair 2008).
As habitats are fragmented, there is less habitat connectivity for animals. When areas are
developed and used for human purposes, the land becomes more fragmented. Humans connect
developed land through roads. For the persistence of endangered species, habitat restoration
needs to be prioritized (Fahrig 1997). Roads have an enormous impact on wildlife communities.
One way to connect dispersed pieces of habitat is by wildlife corridors; channels of intact land
help to increase connectivity (Collinge 2009; Tigas et al. 2002). As corridors aid in creating safe
passage for animals, little is known about how individual animals actually use the corridors to
move between fragments (Tigas et al. 2002). Where natural corridors no longer exist, humans
can help recreate connected landscapes over or under roads in many ways. I will be covering the
creation of habitat connectivity in the mitigation section of the literature review.
Animal Collisions
When habitats are fragmented, they create enormous risks for many animal species. Birds
are highly affected by habitat fragmentation and loss, as there is less connectivity between
parcels of habitat and less space available. As more roads are built and road experience higher
12
volumes of traffic, animals face lower levels of habitat connectivity, increased genetic
bottlenecks, fewer areas to claim as their territories, an increase in traffic noise pollution, and
more chances to be struck by oncoming traffic (Beckman & Hilty 2010; Boves & Belthoff 2012;
Foresman 2004; Husby 2016; Johnson 2005; Loss et al. 2014; Spellerberg 1998; Stewart 2019).
Creating new roads destroys habitat, often forested areas become open spaces, which can affect
the assemblage of species that reside in these newly transformed landscapes. (Benitez- Lopez et
al. 2010; Hovick et al. 2014). The number of animals killed by automobile collisions continues to
grow worldwide (Gunson & Teireira 2015; Newton 1979; Seiler & Heldin 2006).
While the problems roads create for animals are well known, challenges remain with
quantifying these issues and explicitly determining how automobiles affect wildlife (Spellerberg
1998). The majority of the research surrounding animals and vehicles tends to focus on large
hooved mammals due to the effects of ungulates on human lives and the property damage they
can cause (Blackwell et al. 2016). A large portion of the story goes untold when just relying on
roadkill numbers; many interactions and behaviors go unseen when looking at roadkill mortality
events (Clevenger et al. 2003).
There are a number of different survey techniques that can be utilized when exploring
roadkill data. Table 1 (Smith & van der Ree. 2015) displays a variety of survey methods that
compare studies on road’s effects on wildlife.
Some trends have been observed in many studies covering wildlife-vehicle collisions
(WVC). Common spots on roads are known as hotspots that experience many WVC (Clevenger
et al. 2003; Gunson & Teireira 2015; Husby 2016; Taylor 2021). Other factors such as the speed
limit also play a prominent role in where WVC occurs (Foreman et al. 2003; Husby 2017). In
their foundational book, Road Ecology: Science and Solutions, Foreman and colleagues state that
13
vehicles driving faster than 40 mph have a more considerable negative impact on songbirds and
rabbits than vehicles driving slower than 40 mph (pp.120). Birds may not have the sense that cars
pose a significant threat; as DeVault et al. conclude, birds have not evolved to sense vehicles as
threats. Fast-moving vehicles have not been around long enough for birds to be able to recognize
them as predators. The high speeds may overwhelm a bird’s system, so they cannot react in a
timely manner (2014). Road density also is a factor that influences WVC (Clevenger et al. 2003).
Table 1. Survey methods used in studies of road effects on wildlife. For each method type, there are a
number of factors involved that explore how effective the survey method is. These survey types are varied
and range in scale and effort. It is important to recognize that different methods can be useful depending
on the size and scope of a road ecology study. Retrieved from Smith & van der Ree. 2015.
Method
Data
Type
Animal
Spatial
Handling Extent
Resolution/scale Complexity Effort
Roadkill
surveys
Point
No
Smalllarge
Fine
Low
Moderate Low
Animal
tracks
Point,
line
No
Smalllarge
Fine
Low
Moderate Lowmoderate
Camera traps
Point
No
SmallMedium
medium
Low
Low
Wildlife
census:
observational
Point,
line,
area
No
SmallMedium
medium
Low
Moderate Low
Wildlife
census:
interventional
Point
Yes
Small
Moderate
High
High
Animal
tracking
Point,
line,
area
Yes
SmallMedium
medium
High
Lowhigh
Lowhigh
Genetics
Point
Yes/no
Smalllarge
Medium
High
Moderate Moderate
Landscape/
GIS models
Point,
line,
area
No
Large
Coarse
High
Low
Medium
Cost
Low
Low
14
Traffic volume has been cited as having a significant role in WVC by Clevenger et al.
2003 and in other studies: Meunier et al. found that traffic volume was not a factor when raptors
hunted near roadsides. In a hotspot analysis study conducted by Eberhardt et al., they found a
negative relationship between traffic volume and amphibian vehicle collisions. In contrast, they
found a positive relationship between bird collisions and traffic volume (2013).
According to a review that analyzed 49 different studies, mammals and birds had smaller
population sizes close to roads. Mammals were shown to be affected by the presence of roads up
to 5 km away from roads, whereas in general birds could tolerate being closer to roads and were
only affected when they were 1 km or less away from the road. In comparison with other groups
of birds, raptors were found in greater numbers close to roads (Benitez-Lopez et al. 2010).
Roadkill estimates usually undervalue the total number of species killed due to WVC
(Delgado et al. 2019; Jacobson 2005). There are a number of reasons why roadkill counts are
often smaller than the actual amount of roadkill; some surveys do not continue year-round; they
are only funded for specific projects, or animals take roadkill before it is recorded. Eberhardt et
al.’s 2013 study found that 63% of carcasses were gone in a 24-hour window between their
initial roadkill study and their follow-up survey occurring one day later. Guinard et al. conducted
a roadkill survey in southwest France. They compared how long carcasses remained on the
landscape during different seasons; they found that carcasses were picked up or eaten more
quickly during the spring. They attributed this to more scavenger presence during this time of
year. Guinard et al. also found that larger owls stayed on the landscape longer than smaller
songbirds (2015).
15
Lee et al. (2021) further explored ways to account for the low number of found carcasses
by assigning a correction factor that can be used in roadkill studies. For their study on Highway 3
in Alberta, Canada, they applied a correction factor of 2.8 for the detection of ungulates. This
correction factor does not carry over to smaller animals because smaller species do not persist on
the landscape for the same length of time and are generally more difficult to find. Guinard et al.’s
study in France found that they underestimated the percent of owls by 10% and 30% for
songbirds (2015, pp. 100). The survey method also plays a prominent role in detecting roadkill;
more animals are seen if the survey method is done on foot compared to conducting surveys by
car or bicycle (Erritzoe et al. 2003). Guinard et al.’s study in 2015 was conducted by car as well
as on foot. The surveys conducted by moving vehicles underestimated bird carcasses by 33%
(pp.100).
Smaller species make up about 60% of the total kills from WVC; this includes both small
mammals and birds (Seiler & Helldin 2006). Vehicle collisions account for a sizable portion of
bird deaths: Table 2 (Erickson et al. 2005) shows that 80 million birds die each year due to
collisions with automobiles just in the United States, which account for 8.5% of anthropogenic
bird deaths.
16
Table 2. Estimate of yearly bird mortality events in the United States. Retrieved from Erickson et al.
2005.
Mortality Source
Estimated Annual Mortality
% Composition
Buildings
550 million
58.2 %
Power lines
130 million
13.7 %
Cats
100 million
10.6 %
Automobiles
80 million
8.5 %
Pesticides
67 million
7.1 %
Communication Towers
4.5 million
0.5 %
Wind Turbines
28,500
<0.01 %
Airplanes
25,000
<0.01 %
It is important to understand that secondary factors cause birds to be near roadways. For
example, habitat fragmentation has pushed avian species and other wildlife closer to major roads;
predatory birds are drawn to open areas that roadsides provide for hunting (Meunier et al. 2000:
O’Brien 2006). Certain groups of birds are attracted to roads for various resources, including
nesting, hunting, and scavenging. Scavengers, including vultures and eagles, are more likely to
get hit by vehicles (Foreman et al. 2003; Hartley et al. 1996; Husby 2016; Jacobson 2005;
Kociolek et al. 2015; Newton 1979). Scavengers count on roads for their carrion needs and often
use roads to locate carcasses since they are easily spotted from above. This is a universal trend as
studies worldwide have documented scavengers being attracted to roads and consequently killed
due to WVC. Birds that hunt near roads are also susceptible to being hit and killed by vehicles
(Foreman et al. 2003; Jacobson 2005; Kociolek et al. 2015; Lambertucci et al. 2009). Hawks and
owls prey on small rodents near roads and in the median strip of vegetation on highways. Owls
are especially vulnerable while hunting near roads since they are known to fly especially
low. Meunier et al. (2000) detail the importance of roadsides for raptors to hunt, especially in
17
agricultural areas. As more birds reside near roads, the chance of them getting struck by moving
vehicles surges.
In a synthesis of studies by Kociolek et al. (2011), the authors found that birds are likely
to hit struck by water, at lower elevations, and in open areas as opposed to forested habitats.
Birds were also 92% more likely to be hit on raised roads when compared to flat roads
(Clevenger et al. 2003). They also found that birds were more likely to be killed on sections of
road with open spaces rather than forested areas of road. In general, birds are more likely to get
hit during breeding periods (Bujoczele et al. 2011; Foreman et al. 2003; Husby 2016 and
Kociolek et al. 2015). This could be influenced by young dispersing and not being aware of the
danger’s roads present for wildlife.
Birds sampled after being killed due to a WVC were in good health, which went against
what certain studies hypothesized, thinking that birds in poor health would be more likely to get
hit (Bujoczele et al. 2011; Husby 2016; Ramsden 2007). Birds are more likely to suffer injuries
from gusts created by vehicles due to their hollow bones. Orlowski & Seimbieda (2005) noted
that 39% of raptors admitted to the North Carolina Raptor center from 1998-2002 suffered from
injuries due to WVC. The most common affliction was broken bones in the wing region (pp. 15).
Solutions to WVC are being implemented at various scales worldwide. Many of these mitigation
efforts are created for large mammals; mitigation efforts will be discussed in the next section.
One area for improvement with WVC mitigation efforts is studying how birds benefit from these
solutions (Kociolek et al. 2015).
18
Mitigation
Mitigation efforts for WVC range in terms of size, scale, and effectiveness. Certain
projects are centered around promoting practical measures to keep animals off the road, whereas
others target drivers to make them aware of wildlife in the area. In 1992 a survey published by
natural resource agencies in 43 states summarizing mitigation strategies that fell into two
categories: modifying deer behavior and modifying human behavior. The mitigation strategies
that targeted changing deer behavior were wildlife fencing, overpasses and underpasses, hazing,
habitat alteration, and, lastly, mirrors & reflectors. The mitigation strategies that targeted
changing human behavior were public relations, warning signs, warning whistles, highway
lighting, and lower speed limits. The survey was distributed in 43 states. The results quantified
mitigation techniques used in each state and the level of success reported for each mitigation
strategy. Attempts to modify deer behavior were more successful than any of the strategies to
modify human behavior. Wildlife fencing and the installation of overpasses and underpasses
proved to be the most successful. Wildlife fencing was 91% effective, and overpasses and
underpasses were 61% effective. While warning signs were used in almost all 43 states surveyed,
it was less than 10% effective in changing human driving behavior. The most effective strategy
in attempting to modify human behavior was public relations. If outreach is done, drivers can
know more about the potential risks of animals being on roads in certain areas; this can be done
successfully, as discussed in the discussion section.
Distinct strategies have been implemented depending on the targeted species, area, and
cost of the operation. Wildlife mitigation systems, helping to aid animal movement and restore
corridors include vegetated overpasses, open medians, bridge underpasses, culvert underpasses,
19
fences, and detection systems (Cramer & Leavitt 2009, pp. 56). Each mitigation effort listed has
its advantages as well as its drawbacks.
One simple and effective way to make humans aware of wildlife in the area is to install
wildlife signs (Gunson & Teixeira 2015). Signage increases drivers' awareness of animals in the
area that could come onto the road (WSDOT). Signs used to warn drivers about animals on the
site can be broken down into four different categories: standard signs, enhanced wildlife warning
signs, temporal warning signs, and animal detection signs (Huijser et al. 2015, pp, 199). The two
most important factors in determining the success of a sign in alerting humans and making a
lasting impression on are sign location and when the sign in use, both time of day and time of
year, are important (Huijser et al. 2015, pp, 202). Standard signs usually are depicted by a large
picture of an animal without any text. These signs are stationary and do not have a lasting impact
on drivers. Enhanced wildlife warning signs are generally more prominent than the standard
signs and include text or flashing lights. The location of enhanced signs often reflects high
collision zones, and drivers tend to recall these signs for more extended periods than standard
signs. Temporal signage is used at specific times of the year, alerting drivers to certain events
such as an animal migration. These signs have a lasting impact on drivers since they are not
always in use; drivers pay more attention to them. Lastly, live detection signs are only visible
when that animal is in the area. These signs are sometimes connected to radio-collared
individuals and alert drivers when that animal is in the area and could come onto the road.
WSDOT has one of these systems still functioning on U.S. 101 in Sequim to help alert the public
of the elk presence in the area. These systems are not always reliable due to false triggers (sign
flashing when elk are not actually present). It is very taxing to keep up with attaching radio
collars to individual elk (WSDOT 2021).
20
One example of success with temporal signs as well as radar speed limit signage is a
project based in Jackson Hole, Wyoming. The project is called Give Wildlife a Brake. The
project’s effort is centered around alerting the public to slow down and pay attention to large
animals that could be present. The project has four stationary radar signs and three dynamic
message signs, like the sign from WY 360 shown in Figure 4. These efforts have already
resulted in a decrease in animals hit and killed in the area: from 2010-2014, 36 moose were
struck and killed on WY 360, while in 2015, no moose were reported as killed (Jackson Hole
Wildlife Foundation n.d.).
Figure 4. An example of a temporal dynamic sign that is used in the program, Give Wildlife a Brake.
Retrieved from The Jackson Hole Wildlife Foundation’s website.
Lowering traffic speeds and centering traffic on fewer, more commonly utilized roads are ways
to help alleviate problems caused by WVC as well (Forman et al. 2003; Gunson & Teixeira
2015; Kociolek et al. 2015).
Wildlife crossing structures are well-known wildlife mitigation efforts engineered to
assist terrestrial wildlife in crossing over or under roadways. Wildlife crossing structures have
been built all over the globe. The first crossing structure built in the United States was in Utah in
1978. The crossing structure was implemented to aid in the migration of deer. In the 1980s and
21
1990s, a considerable effort went into creating habitat connections in the Everglades. State and
federal partners helped oversee a project consisting of 23 underpasses and bridges. Their main
focus was to lessen the number of Florida panthers killed due to WVC and help enhance water
movement for the American alligator. (Foreman et al. 2003). As of 2009, more than 700 wildlife
crossing structures exist in North America, as well as thousands of crossing structures to aid
aquatic wildlife (Cramer & Leavitt 2009).
In Washington state, WSDOT works with partners such as the Washington Department
of Fish and Wildlife to determine where hotspots of WVC occur. WSDOT has created a number
of wildlife crossings, including the Interstate 90 project, which features a vegetated over-crossing
and an undercrossing from Hyak to Easton, elk are using the overpass frequently and species as
cougars have been seen on the undercrossing. A great horned owl was documented landing on
the camera on the overpass on I90. Another successful project is on U.S. Highway 97; Janis
bridge undercrossing was retrofitted in 2019, the vegetation was cleared from the structure, and a
mile of wildlife exclusion fencing was built. This area is one of the top deer collision areas in the
state, with 350 deer being hit each year over a twelve-mile stretch. Since 2019, the crossing is
seeing, on average, six mule deer per day. Other faunas such as ring-necked pheasants, raccoons,
and bobcats have been documented using this passage as well (Conservation Northwest 2019,
WSDOT 2021).
Birds often get overlooked when it comes to crossing structures, and there have been very
few papers published on crossing structures for avian species (Pell & Jones (2015). What is
known about the relationship between wildlife crossing structures and birds is that even though
they are not commonly built with birds in mind, birds benefit from them (Jacobson 2005; Jones
& Bond 2010; Pell & Jones 2015). To increase the effectiveness of wildlife crossing structures,
22
adding wildlife fencing to these areas and jumpouts and/or soil berms help keep animals out of
the roadway. Eight-foot-tall wildlife exclusion fencing is a beneficial way to keep animals off of
roads. When fencing is combined with wildlife crossing structures, it decreases the chances of
animals finding a way onto the road (Huijser et al. 2010; WSDOT 2021). A great way to monitor
the long-term use of wildlife crossing structures is to install wildlife cameras. Data from wildlife
cameras help understand what animals are using the cross structures, what animals are being
repelled from such structures, and identify wildlife movement patterns (WSDOT 2021).
Mitigation efforts that separate birds and other species from roads as early as possible in
the planning and construction efforts will be most effective long term (Jacobson 2005; Jones &
Bond 2010). Mitigation projects should also be widespread initiatives to account for the mobility
of birds (Kociolek et al. 2011). Some measures that help separate birds from roads include
removing roadkill promptly not to attract scavengers to the roadside. HawkWatch International
conducted a study to better understand how eagles were using roads to scavenge on roads. They
found that moving roadkill ten meters off the road significantly decreases the chances of eagles
flushing into traffic when cars pass (Taylor 2021). It has been found that tall barriers on either
side of a road will encourage birds to fly across the road without dipping down to the height of
the road. The birds will choose to fly within their line of sight straight across the road (Delgado
et al 2019; Pons 2000). Placing poles by the roads mimics the appearance of a physical barrier,
so birds are not tempted to fly into roadways. (Kociolek et al. 2015; Jacobson 2005). Table 3
highlights mitigation strategies surrounding differing communities of birds that interact with
roads.
23
Table 3. A visual representation of different problems roads create and solutions to fit the needs of
particular groups of birds. Retrieved from Jacobson, 2005.
Group
Impacted
Problem
Suggested Solution
Walking
Birds
Non-flying birds incur great mortality risk.
Crossing structures with large openness ratios
(underpasses) or wildlife over-crossings.
Water Birds
Winds over bridges can slam flying birds
into vehicles.
Diversion poles on bridge decks.
Owls
Owls hunt at headlight height.
Diversion poles or short fences along highway
medians and right-of-way.
Ground
nesters
Mowing right-of-way kills nesters.
Mow after August 1.
Scavengers
Corvids or raptors are killed while foraging
on roadkill, attracted scavengers reduce
productivity adjacent to highways.
Reduce roadkill & remove roadkill from road.
Migrant
landfalls
Exhausted cross-gulf migrants fly into
vehicles.
Low temporary fences to encourage higher
flight across roads.
Frugivores
Fruiting median plants attract birds across
traffic
Plant non-fruiting varieties & remove fruiting
varieties.
Winter
finches
Deicing salt or sand attracts birds to road
surface.
Velocity spreaders, road temperature sensors to
reduce quantities, concentrate runoff
appropriately, & public education program.
Long-term monitoring efforts should be implemented to help better understand what
makes a successful mitigation project. Cramer and Leavitt (2009) suggest monitoring a crossing
structure for a minimum of three years after it has been built; most wildlife will not use a
structure for about two years, this may vary due to animals that are not always present in the area
that may only use it once or twice a year during its migration. It is an exciting time for change as
much of our highway infrastructure needs to be updated in the next few decades; there is room to
retrofit bridges and culverts to create wildlife corridors (Cramer & Leavitt 2009).
24
Methods
The main objective of this research was to identify the relationship between locations of
reported raptors collisions on Washington State Highways. Raptor collision information was
extracted from the Washington Department of Transportation’s (WSDOT) Wildlife Carcass
Removal Database (WCRDB). Maps were created in ESRI’s ArcGIS Pro application utilizing a
series of GIS modeling steps.
Study Area & Data Sources
The study area included Interstate and state highways within Washington State (Figure
5). The data used in this study came from the Washington Department of Transportation’s
Wildlife Carcass Removal Database (WCRDB). A disclaimer about the WCRDB from WSDOT:
This data is protected under the same United States Code (Section 409 of Title 23) as the
WCR data. Any collision data furnished is prohibited from use in any litigation against
state, tribal or local government that involves the location(s) mentioned in the collision
data.
The data in the WCRDB has been collected for almost fifty years, starting in 1973. When the
database was first established, maintenance personnel collected data using a paper form, as
shown in Figure 6. Data collection has changed and evolved alongside advances in technology.
Personal digital assistants replaced the paper wildlife roadkill report form in 2010, and in 2015,
iPads were implemented to collect data in the field.
25
Figure 5. A map of the study area, Washington State Highways shown in
pink. Created in ArcGIS Pro.
Figure 6. An example of the roadkill spreadsheets before the data was collected by iPad.
Retrieved from WSDOT.
26
The roadkill data is reported initially in the Highway Activity Tracker by WSDOT
maintenance staff, and then digitally transferred to the WCRDB. Once uploaded in the WCRDB,
a member of the habitat connectivity team processes each individual record. If there are errors
within a record, the maintenance personnel who entered the record is contacted by email (if
necessary). A habitat connectivity team member fixes each error manually. The most common
errors include wrong location and incorrect species identification (Croston pers. comm).
In order to separate raptor data from the rest of the records in the WCRDB, I exported all
accounts of bird carcasses to an Excel spreadsheet. I only exported data from 2015-2020 as it
was all collected using iPads. From here, I filtered the data to display all raptor kills. This
included bald eagles, golden eagles, hawks, red-tailed hawks, owls, and turkey vultures in the
database. I then created a separate spreadsheet for each group of raptors. The fields collected in
the WCRDB are as follows, removal date, species, region (Eastern EA, North Central NC,
Olympic OL, South Central SC, Southwest SW, and Northwest NW), State Route, Milepost,
Latitude & Longitude, Type of Animal, Sex-if known, Age-if known, disposal method, as well
as the observers' name. Figure 7 below displays the different regions within Washington that
WSDOT uses.
27
Figure 7. A map displaying the different regions of Washington, a field that is collected in the wildlife
carcass removal database. Retrieved from WSDOT.
I looked through every raptor carcass in the system to double-check its authenticity based
on location (both regionally as well as the Lat/Long coordinates). I had to manually input the
latitude and longitude for many raptor records in the system because only their mileposts and
highway numbers were put into the records. In order to locate the specific mileposts, I utilized
WSDOT’s Milepost Values map image layer via the ArcGIS online platform. WSDOT’s
Milepost value layer is split up into 1/10 of a mile segments. WSDOT’s Milepost Values layer
was created on August 13, 2014, and last updated on February 2, 2020. After locating each
reported raptor’s location, I then looked up that specific spot in Google Maps. After finding the
raptor carcass collision location in Google Maps, I was able to copy and paste the latitude and
longitude into the existing raptor Excel spreadsheets.
28
I also looked up all raptor identifications in the Highway Activity Tracker (HATS),
where the record could have an attached photo(s). The images help to reinforce a positive
identification. I also wanted to check to see if I could identify owl and hawks to species by
looking up their incident numbers. HATS does not have an option to identify owls or hawks
(besides red-tailed hawks) to species. Posters have been created by the habitat connectivity
biologist and distributed to the maintenance departments around the state in order to help make
the correct identification. The owl poster is shown below in Figure 8. The owls displayed on the
poster make up a collection of both common owls and rare owls (Northern Spotted Owl). Even
though there is not an option in the pull-down menu of species, the type of owl can be included
in the comments field in the HATS database.
Figure 8. Owls in Washington state, a key to help identify owl carcasses. Created by Habitat Connectivity
Biologist Glen Kalisz, Retrieved from WSDOT.
29
While looking up species in the HATS database, I corrected a number of misidentified
raptors. Besides the records I was able to correct, many of the records did not have photos
attached, and there was no way for me to confirm these raptor identifications. Using Excel, I
created a spreadsheet for each group of raptors as well as a separate spreadsheet for all combined
raptor records.
ArcGIS Pro Applications
I uploaded the different raptor data as CSV files into ArcGIS Pro to create a series of
maps and graphs. I used an elevation raster layer from the USGS. I downloaded a highway
shapefile layer from WSDOT. Maps were created using the WGS 1984 coordinate system.
Kernel Density Analysis
To determine which sections of the highway system account for the highest frequencies
of reported raptor collisions, I ran a kernel density analysis in ArcGIS Maps. Kernel density tests
determine “the density of point features around each output raster cell.” (ArcGIS Pro). The
kernel density calculation is shown in (Figure 9, ESRI, n.d.).
Figure 9. The expanded equation to calculate kernel density, Retrieved from ESRI, n.d.
I made the output size of the cell 30 m (98.4252 ft) to be consistent with other spatial
analyses. I also made the diameter 10,720 ft, the same extent that the Washington Department of
30
Transportation uses to catalog deer collision hotspots in the state. I measured the distance from
the beginning to the end of each hotspot location using the ‘map an SR (state route)’ tool
available in WSDOT’s GIS toolbox.
ArcSurvey 123 Application
I also created a new tool for raptor carcass data collection. I designed a data entry system
using the ArcGIS Survey123 platform. The survey has a more comprehensive list of raptor
species; the user can select the type of raptor and then is given a drop-down menu of species
within that group. This raptor list is located in Appendix B. The survey design is meant to be
used as a collective science effort to strengthen raptor collision data on a state level. Using
Survey123, I developed the schema of the application in an Excel spreadsheet.
Limitations
Numerous limitations come with using the WCRDB. One inconsistency in the data is the
different versions of data collection methods as the database has evolved over time. Another
limitation of this dataset is precision: the location of the carcasses is not calculated using GPS
data but instead relies on milepost markers, which can be imprecise especially if milepost
markers are missing in certain areas.
Another drawback of the WCRDB are the taxonomic distinctions available within the
database. In particular within raptors, owls cannot be broken down into types of owls, same goes
for hawks with the exception of red-tailed hawks. There is a way to upload photos, but these can
only be accessed by the HATS database, which is a separate system altogether, and pictures are
not required.
31
This database does not account for animals struck by a vehicle that then leave the scene
before dying. Carcasses could also be picked up and moved before WSDOT is aware of them,
affecting the total number of raptors killed on Washington State Highways. The quality and
quantity of entries in the WCRDB can vary from region to region and from the individual
maintenance employee.
Regardless of these limitations, having a database that provides information on raptor
collisions in Washington State does help to visualize where they occur. This data does not exist
in any other database on such a widespread level within the state of Washington.
32
Results
The total number of reported raptor carcasses from 2015-2020 is reported in Table 4.
Owls account for 68.7% of the data. The raptor groups recorded in the WCRDB were bald
eagles, golden eagles, turkey vultures, hawks, and owls. Table 5 contains the reported raptor
collisions by year and species/group. Table 6 displays the owl carcasses by month; each month
recording is a calculation of all owl carcasses reported for that month between 2015-2020. Table
7 contains all owls from the HATS database identified to species by photo or written in
comments. 154 of the 683 records contained either photographic or verbal identification. These
owl records are also broken up by region in the state.
Table 4. The total recorded raptor carcasses in the WCRDB from 2015-2020, broken up by whole number
as well as percent.
Total Carcasses Reported 2015-2020
% of the Total
Bald Eagle
18
1.8
Golden Eagle
7
0.7
Turkey Vulture
9
0.9
Owl
683
68.7
Hawk
278*
27.9
Total
995
100
Type of Raptor
* 62 identified as Red-tailed Hawks
33
Table 5. Reported raptor carcasses in the WCRDB from 2015-2020, broken up into raptor group and
separated by year.
Bald Eagle
Golden Eagle
Turkey Vulture
Owl
Hawk
Grand Total
2015
2
1
2
36
21
62
2016
1
0
1
102
36
140
2017
8
2
0
128
49
187
2018
3
0
0
155
48
206
2019
2
1
4
156
55
218
2020
2
3
2
106
69
182
Total
18
7
9
683
278
995
Table 6. Reported owl carcasses in the WCRDB from 2015-2020, as monthly totals.
Month
Number of Owl Carcasses Reported
January
66
February
41
March
45
April
37
May
38
June
30
July
35
August
55
September
53
October
101
November
94
December
88
34
Table 7. A display of all owl records with species written in the comments section of the WCRDB from
2015-2020, by WSDOT Region (see Figure 7).
Owl Species
EA
NC
NW
OL
SC
SW
Total
Barn Owl
1
19
19
10
13
9
71
Great Horned Owl
11
9
3
3
10
9
45
Barred Owl
1
1
6
9
1
18
36
Northern Saw-whet Owl
0
0
0
0
0
2
2
Region Total
13
29
28
22
24
38
154
Figure 10 illustrates raptor type by region of the state. The northwest region has the most
total raptor carcasses in the state. The eastern, northwest, and south-central region each have
more than one hundred reported owl carcasses. Figure 11 depicts all reported raptor collisions
by county. Skagit County has the most collisions in any particular county.
Figure 10. Raptor types reported in each region of the state in the WCRDB from 2015-2020.
35
Figure 11. A graphic display of raptor collisions by county reported in the WCRDB from 2015-2020.
Figures 12-14 show the locations of carcass data along highways in Washington State
(all reported, owls and hawks). There is a significant discrepancy between the records with
photos attached versus records without photos from the HATS database.
See what you think of the 2 images on one page (next page) – if you like the larger
images better, go ahead and increase their size again. I just think it’s nice to be a little more
compact if possible.
36
Figure 12. All reported raptor carcasses displayed spatially from 2015-2020.
Map was created in ArcGIS Pro.
Figure 13. A spatial display to document the locations and frequencies of owl records
with and without photos attached in the HATS database. Map was created in ArcGIS Pro.
37
Figure 14. A spatial display to document the locations and frequencies of hawk records with and without
photos attached in the HATS database. Map was created in ArcGIS Pro.
Figure 15 shows the locations of hotspot areas in the state as you can see reported raptor
collisions occur in specific areas in the state, they do not occur evenly across state highways.
Figures 16, 17, 18, and 19 display the zoomed-in areas of each of the four hotspot areas for
reported raptor collisions. Figure 16 is a hotspot just north of Vancouver Washington in an area
featuring hay/pastureland type as well as human development. Figure 17 displays the second
hotspot on Interstate 5, this is where the most highly concentrated area of reported raptor
collisions occurs. Figure 18 shows the first hotspot location on Interstate 90, this hotspot is in
Grant county and the land type in this area is mostly used for cultivated cropland. Figure 19
38
displays the second hotspot on Interstate 90, southwest of Spokane. The reported raptor
collisions in this area occur remarkably close together.
Figure 15. Kernel Density distribution of raptor collisions in Washington. Map was created in ArcGIS
Maps.
39
Figure 16. A close-up view of raptor the I5 raptor hotspot A. Map was created in ArcGIS Pro.
40
Figure 17. A close-up view of raptor the I5 raptor hotspot B. Map was created in ArcGIS Pro.
41
Figure 18. A close-up view of raptor the I90 raptor hotspot A. Map was created in ArcGIS Pro.
42
Figure 19. A close-up view of raptor the I90 raptor hotspot B. Map was created in ArcGIS Pro.
43
Discussion
The questions I set out to answer through this research were:
1. Where are raptors being hit and killed due to vehicle collisions on Washington State
Highways?
2. What are the emerging temporal and spatial trends in the data? Are there hotspots that
appear? Are there certain times of year in which more raptor collisions are reported?
3. What is the methodology in capturing raptor data within the wildlife carcasses removal
database? Are there gaps in the data? How can this data source be strengthened?
I was able to answer each of these questions. For question 1, the state's northwest region
accounts for 30.25% of all raptor collisions in the state from January 1, 2015, to December 31,
2020. The hotspots in the state are situated in areas that experience high traffic volumes. This
does not mean that there are necessarily more raptors in these areas, just that the raptors that are
present are more likely to be hit and reported due to the number of vehicles on the roads.
I would have expected more scavenger presence in the overall number of raptor carcasses
recorded. The literature cites that the frequency of scavengers involved in WVC is higher than
other groups of birds (Foreman et al. 2003; Hartley et al. 1996; Husby 2016; Jacobson 2005;
Kociolek et al. 2015; Newton 1979). In a presentation given by HawkWatch International, they
reported eagle carcasses being picked up and stolen before transportation departments could pick
up and document those individuals being hit and killed by vehicles. They were able to obtain this
data by cameras that were monitoring other roadkill carcasses that would attract both bald and
golden eagles (Taylor, 2021). Compared with different raptor types, the proportion of owls is
high, which I would anticipate since they fly low and are attracted to roads during crepuscular
and nocturnal times, making them difficult for drivers to see.
44
Large owls make up most owls recorded, with species noted in the comments field of the
WCRDB. Large owls would likely be found more frequently since they are easier to spot than
smaller owl species. I was surprised that two saw-whets were identified (both accounts had
photos attached) due to their small stature. Their carcasses could have easily been carried away
in the grill of a vehicle or blown off the road completely. A 10-year study conducted in Cape
May, New Jersey, found 250 raptor carcasses over a 145 km route. Twelve different species of
hawks and owls were found (6 species of owls, six species of hawks). Owls comprised 88% of
the data, with saw-whets making up 52% of the total owl carcasses. Loos and Kerlinger (1993)
reported that 87% of saw-whets and 72% of eastern screech owls were hit between November
and January. They hypothesized that many of the owls hit during this period were dispersing and
were last year's young—this study is unique due to the small owls (saw-whets and eastern
screech owls) making up most of their data (Loos and Kerlinger 1993). The most common owl
cited in the literature for being involved in WVC is the barn owl (Boves & Belthoff 2012;
Guinard et al. 2012; Meunier et al. 2000; Ramsden 2007). Boves and Belthoff (2012) found barn
owls to be the most common species hit between July 2004 and June 2006 on I-84 in Idaho. Of
63 species they encountered, barn owls accounted for 32% of all roadkill mortality events. I also
found barn owls to be the most prominent owl in the reports; Table 7 shows that 71 out of the
154 owls with species listed were identified as barn owls.
To answer question 2, regarding temporal trends, I found similar chronological patterns
as Loos and Kerlinger (1993) within the owl data; 51% of owls were hit between October and
January, as shown in Table 6. Over the five years of data collection, 101 owls were hit and killed
in the month of October, which contrasts with Loos and Kerlinger (1993) as they do not cite
October as a month that their study saw significant raptor records.
45
Increased traffic volume is known to play a role in raptor collisions. A fifteen-year study
from the Barn Owl Trust in Britain found a positive correlation between traffic volume and the
number of barn owls being struck (Ramsden 2007). This becomes an important factor on major
roads; The Barn Owl Trust calculated that Barn owls are three times more likely to be involved
in a WVC on main roads than to be seen alive. On smaller roads in Britain, they found that Barn
Owls were fifty-seven times more likely to be found alive rather than dead (Ramsden 2007).
Washington state has experienced a significant increase in the annual vehicle miles traveled per
year. Figure 20 (2016 WSDOT) shows the increase from 1990 to 2016 in both urban and rural
traffic volumes; this information came from the 2016 Annual Traffic Report published by
WSDOT. Appendix A displays this information in a table format displaying the changes in
percent from year to year.
Figure 20. The amount of vehicle miles traveled (AVMT) in Washington state from 1990-2016. Data is
separated by rural and urban miles. Graph is taken from 2016 Annual Traffic Report, Washington
Department of Transportation, pp.55.
46
Hotspots in Washington
To answer question 2 with regards to spatial trends, I found that four major hotspot areas
emerged for raptor collisions in the state. 239 out of the total 995 strikes occurred on Interstate 5,
comprising 24% of the data. 78 of the reported raptor collisions were between mileposts 13-76.
The land cover in this area is hay/pastureland as well as developed land. These mileposts span
across three different counties: Clark County, Cowlitz County, and Lewis County. 151 of the
raptor collisions on Interstate 5 took place between mileposts 167-274. The land cover in this
area is developed, hay/pastureland, and cultivated cropland. These mileposts span across four
different counties, starting in King County, crossing Snohomish and Skagit county, and ending in
Whatcom County.
Skagit county had the most raptor collisions of any individual county in the state. The
Skagit Valley is home to very fertile soil, and many farming practices occur here. The Skagit
area hosts a rich estuarine environment as well. Due to the area’s wealth of resources, it attracts
many bird species. The Skagit is an important migration area for many avian species, including
raptors, shorebirds, swans, and snow geese (Audubon n.d.; Birds of Winter 2018). One specific
location known as the Skagit River Delta is a winter migration hotspot for over 50,000 snow
geese, and 262 individual species of birds have been identified via Ebird here (Birds of Winter
2018).
125 reported raptor collisions from 2015-2020 occurred on Interstate 90. This makes up
12.5% of the total data. 49 of these strikes took place between mileposts 138-175. The main
landcover in this area is cultivated cropland and lies entirely within Grant county. The other
hotpot on I90 is between mileposts 220-298; 56 strikes were reported here. The landcover types
47
in this area are cultivated land, herbaceous land, and developed land. These mileposts are in
Adams and Spokane county.
Gaps in Current Research
Roads and their effects are widespread; the research in road ecology needs to reflect the
size and magnitude of its subject. Systematic methods to detect roadkill mortality rates need to
be implemented at a larger scale. Local efforts use state, region, and community science data to
understand WVC rates better. These WVC datasets, however, are usually need-based and not
sustained efforts. The long-term effects of roads need to be better monitored (Beckman & Hilty
2010; Forman et al. 2003; Spellerberg 1998). Due to the lack of widespread WVC data collection
efforts, the mortality rates of animals are probably more significant than what estimates currently
are (Foreman et al. 2003; Jacobson 2005). The parties responsible for collecting and reporting
carcasses need specialized training based on bird identification so they can recognize the
different species of birds that are getting hit and killed in vehicular collisions (Ramsden 2007).
There is a gap in the knowledge base surrounding the effects of birds and roads; more research
and understanding need to focus on birds and roads to help mitigate the negative repercussions
roads present to the avian community.
Future Studies
To answer my third and final question, with regards to how can this data source be
strengthened, I began to think about different ways the data could be collected and improved
upon. The limitations listed in the methods section display the finite species identification
options in the WCRDB when it comes to birds, in particular raptors. Along with the shortage of
48
avian options, another challenge is the scarcity of photos captured to help strengthen raptor
identification.
To create an effective roadkill carcass survey, there are four objectives to consider, as
outlined in (Figure 21, Shilling et al. 2015).
Figure 21. A systematic way to select objectives of a WVC survey. Figure 62.1 Shilling et al. 2015.
When creating a wildlife collision survey, it is important to decide what objective(s) are
most important to the study; this will influence how data is collected. Shilling et al. 2015 also
depict two different strategies for collecting data. Studies can utilize opportunistic and random
observations or transect/targeted route observations. The opportunistic and random observations
rely on people reporting WVC when they happen across carcasses. The latter method is based on
people following specific routes actively looking for carcasses. Both methods are beneficial in
collecting carcass observations.
49
One effort that has bolstered carcass data specific to ungulates in WSDOT’S WCRDB is
the salvage data system. The salvage database is a partnership between WSDOT and the
Washington Department of Fish and Wildlife. This system is a way to maintain and keep track of
all deer and elk carcasses salvaged in Washington. The salvage system began in 2016 and has
given the WCRDB a wealth of data since then. The salvage removal system is a form of
community science because community members are helping to identify where deer and elk are
hit and removed in the state. I believe a community science effort surrounding raptor carcass data
would help account for raptors all over the state, not just those hit on Washington State
Highways.
I created a survey using ArcGIS 123 that could be used as a data collection tool to gather
information from many different sources. This would help to strengthen the raptor WVC data in
the state. I chose ArcGIS Survey 123 as the application for its user-friendly features. If a roadkill
survey is too complicated, the results can be erroneous (Kinyon, 2017). The survey has options
to choose from thirty-three different raptor species found in the state of Washington, some more
commonly found than others. The thirty-three species it accounts for are listed in Appendix
B. The survey allows the user to take photos and provide their GPS location. It also has an option
for their level of certainty that they identified the species correctly. The survey lets the user
select what type of road they were on and the speed limit where they found the carcass. The
survey also allows the user to estimate how long they believe the raptor has been dead. These
questions would help provide more data about raptor collisions in Washington State outside
those made by maintenance personnel in the HATS database. This would allow more questions
to be asked and answered about the relationship between roads and raptors, especially
concerning certain species and hotspot locations that do not occur on state highways.
50
Numerous community science efforts focus on reducing animal collisions on roads and
increasing local knowledge. The more people aware of high-use areas from wildlife, the better
they can prepare when driving through those areas. Two examples of successful community
science projects both took place on Highway 3 in Alberta, Canada, in the Crowsnest Pass area.
Road Watch was a productive community science effort that relied on individuals driving and
reporting any animal they saw near the road (alive or dead) using a web-based mapping
application (Lee et al. 2010; Paul 2002). This project had over 4,044 reported wildlife
observations. It succeeded in its goal of engaging the public in data collection. In 2007, they
surveyed to see if community members who participated in the data collection felt they had
gained more knowledge about WVC. 85% of survey participants agreed that their knowledge of
WVC had increased through the Road Watch project (Lee et al. 2010).
The later study conducted in a partnership with a highway maintenance department
called Citizen Count ran from 2014-2017. The aim of the community science project was to
collect roadkill information pre- and post-construction on a series of mitigation projects,
including an underpass, wildlife exclusion fencing, and jumpout installation. The volunteers
walked the same transect each week from 2014-2017. They were taught how to record data using
an app called Collision Count, and over twenty volunteers participated in the project. The
maintenance contractors also surveyed Highway 3 twice a day from 1997-2017(Lee & Rondeau,
2018). The combination of maintenance crews and community members working together
creates a wealth of local knowledge about an area. It helps to engage the surrounding
communities to be more aware of the roadkill issues, resulting in drivers being more alert to
animal movements.
51
With an increase in data surrounding wildlife collisions, more time and energy can go
into fixing WVC problems. There are numerous ways to help wildlife navigate our infrastructure
and be successful in the wild. Community science efforts could help pinpoint where raptors are
being hit in Washington; this could solve other challenges that wildlife face in these areas.
For the purposes of my study, I looked at all raptor vehicle collisions in the state; future
studies may decide to look at the effects of roads on individual species or groups of raptors (e.g.,
owls). Since not all raptors occupy the same scale, looking at individual species may be
beneficial in better understanding individual species' spatial ranges in union with the hotspots
(Sevigny et al. 2021).
If particular species are being hit and killed by vehicles more often, this could lead to
long-term changes in a population; these changes over time could impact an ecosystem (Santos
et al. 2016: Seiler & Helldin 2006). It would also be interesting to study individual raptor species
to see how certain species better adapt to roads over time, understand the relationship between
roads and cars, and use roads to their advantage.
One other future study area that I think would be influential in understanding raptors,
open agricultural land, and the relationship of the road would be to look specifically at the Skagit
Valley region. As noted earlier, this area is home to fertile fields, numerous migratory birds, and
an estuarine environment. Skagit county also experienced the most raptor vehicle collisions than
any other county from 2015-2020. A systematic road carcass system would capture more
information in the area. Getting in touch with local wildlife rehabilitators and rehabilitation
centers would also help account for raptors not killed on site. This would be a way to capture
data concerning injured raptors as well.
52
Conclusion
Animal collisions only tell a small part of the story in the realm of road ecology. There is
so much more to learn about the complex relationship between animals and roads. Raptors are a
unique set of birds to study; roads are important for hunting grounds and scavenging resources.
As long as raptors find the benefits of roads to overshadow the potential danger of vehicular
collisions, they will continue to utilize these resources.
Road ecology is a blossoming branch of science full of opportunities. With so many local
efforts to increase habitat connection for animals, the future is bright. Transportation planners
should consider data from WVC to make informed planning decisions. Road ecology lends itself
to collaboration between a suite of different professions. Creating and restoring habitat
connectivity is vital for the well-being of animal communities. As we look forward, many roads,
bridges, and other forms of infrastructure are outdated and currently being worked on or will be
in the near future. The time is right to update antiquated designs to incorporate mitigation efforts
for wildlife.
The trends found in my research are capable of aiding in creating mitigation efforts to
help lessen raptor strikes, especially in the hotspots of collisions. The WCRDB can be
strengthened when it comes to raptor and bird identification. This work is a great baseline to
strengthen in the future.
Moving forward, I hope to educate people and make them aware of potential animals on
roads, whether they are fly, walk, slither, or hop across the street. Educating the public on road
awareness can not only benefit wildlife but drivers as well. Community science is a powerful
tool that can help in collecting data and increasing awareness of WVC. I also hope people who
53
read this work will become more aware of the challenges that animals face due to anthropogenic
causes and feel empowered to become involved in community science efforts.
54
Bibliography
Audubon. (n.d.). Important Bird Areas, Skagit Bay Washington. Retrieved from
https://www.audubon.org/important-bird-areas/skagit-bay
Beckman, J.P. & Hilty, J.A. (2010). Connecting Wildlife Populations in Fragmented
Landscapes. In J.P. Beckmann, A.P. Clevenger, M.P. Huijser & J.A. Hitly, (Ed.)
Safe Passages: Highway, Wildlife, and Habitat Connectivity (pp. 277-292).
Washington, DC: Island Press.
Benitez-Lopez, A., Alkemade, R., & Verweij, P.A. (2010). The impacts of roads and
other infrastructure on mammal and bird populations: A meta-analysis. Biological
Conservation, 143, 1307-1316.
Birds of Winter (2018). A Skagit Valley Experience. Retrieved from
http://birdsofwinter.org/
Blackwell, B., F., DeVault, T., L., Fernandez-Juricic, E., Gese, E.M., Gilbert-Norton, L.,
& Breck, S.W. (2016). No single solution: application of behavioural principles in
mitigating human-wildlife conflict. Animal Behaviour, 120, 245-254.
Bloom, P.H., & McCravy, M.D. (1996). The Urban Buteo: Red-Shouldered Hawks in
Southern California. In D.Bird., D. Varland & J. Negro, (Ed) Raptors in Human
Landscapes (pp. 31-39). London, England: Academic Press.
Boves, T. J. & Belthoff, J. R. (2012). Roadway Mortality of Barn Owls in Idaho, USA,
The Journal of Wildlife Management, 76 (7), 1381-1392.
55
Bujoczek, M., Ciach, M. & Yosef, R. (2011). Road-kills affect avian population quality.
Biological Conservation, 144,1036-1039.
Burfield, I.J., (2008). The Conservation Status and Trends of Raptors and Owls in
Europe. Ambio, 37 (6): 401-407.
Canada, Government of Alberta. (2002). Owls and the food chain (pp. 1-3). Retrieved
from https://open.alberta.ca/dataset/a6770937-8705-46b5-966dcd4a16dc1013/resource/3b9dea09-7629-4e86-96ff-fa89ea232ee0/download/owlsfoodchain-jul17-2009.pdf
Clevenger, A.P., Chruszcz, B. & Gunson, K.E.( 2003). Spatial patterns and factors
influencing small vertebrate fauna roadkill aggregations. Biological
Conservation, 109, 15-26.
Collinge, S.K. (2009). Ecology of Fragmented Landscapes. The Johns Hopkins
University Press.
Conservation Northwest. (2019). Safe Passage 97 project moving forward
despite setbacks from Olympia. Conservation Northwest.
https://www.conservationnw.org/news-updates/safe-passage-97-moving-forward/
Cramer P. & Leavitt, S. (2009, Winter). Right of way: giving animals safe passage
across roadways. The Wildlife Professional, 56-60.
56
Delgado, J.D., Humia, J, D., Pereiras, A. R., Rosal, A, Palenzuela, M, D, V., Morelli, F.,
Hernandez, N. L. A. & Sanchez, J. R. (2019). The spatial distribution of animal
casualties within a road corridor: Implications for roadkill monitoring in the
southern Iberian rangelands. Transportation Research Part D, 67, 119-130.
DeVault T. L., Blackwell B. F., Seamans T. W., Lima S. L., Ferna´ndez-Juricic E.
(2014). Effects of Vehicle Speed on Flight Initiation by Turkey Vultures:
Implications for Bird-Vehicle Collisions. PLOS ONE, 9(2): 1-9.
Donázar, J. A., Cortés-Avizanda, A., Fargallo, J. A., Margalida, A., Moleón, M.,
Morales-Reyes, Z., . . . Serrano, D. (2016). Roles of Raptors in a Changing
World: From Flagships to Providers of Key Ecosystem Services. Ardeola, 63(1).
Eberhardt, E., Mitchell, S., & Fahrig, L. (2013). Road Kill Hotspots Do Not Effectively
Indicate Mitigation Locations When Past Road Kill Has Depressed Populations.
The Journal of Wildlife Management, 77(7): 1353-1359.
Erickson, W. P., Johnson, G. D., & Young, D.P. Jr. (2005). A summary and comparison
of bird mortality from anthropogenic causes with an emphasis on collisions.
USDA Forest Service Gen. Tech. Rep, 1029-1042.
Erritzoe, J. Mazgajski, T. D., & Rejt L. (2003). Bird casualties on European roads-a
review. Acta Ornithologica, 38 (2): 77-93.
Fahrig, L. (1997). Relative Effects of Habitat Loss and Fragmentation on Population
Extinction. The Journal of Wildlife Management, 61(3), 603-10.
57
Fahrig, L. (2003). Effects of Habitat Fragmentation on Biodiversity. Annual Review of
Ecology, Evolution, and Systematics, 34, 487-515.
Forman, R.T.T. & Alexander, L.E. (1998). Roads and Their Major Ecological Effects.
Annual Review of Ecology and Systematics, 29, 207-231.
Forman, R. T. T., Sperling, D., Bissonette, J.A., Clevenger, A.P., Cutshall, C. D., Dale
V.H., Fahrig, L., France, R., Goldman, C. R., Heaune, K., Jones, J. A., Swanson,
F. J., Turrentine, T. and Winter, T.C. (2003). Road Ecology: Science and
Solutions. Island Press.
Guinard, E., Julliard, R., & Barbraud, C. (2012). Motorways and bird traffic casualties:
Carcasses surveys and scavenging bias. Biological Conservation, 147, 40-51.
Guinard, E., Prodon, R., & Barbraud, C., (2015). Case Study: A robust method to obtain
defendable data on wildlife mortality. In R. van der Ree, D.J., Smith, and C.
Grilo, (Ed.) Handbook of Road Ecology (pp. 96-100). John Wiley & Sons, Ltd.
Gunson, K. & Teixeira, F.Z. (2015). Road-wildlife mitigation planning can be improved
by identifying the patterns and processes associated with wildlifevehicle collisions. In R. van der Ree, D.J., Smith, and C. Grilo, (Ed.) Handbook of
Road Ecology (pp. 101-109). John Wiley & Sons, Ltd.
Hartley, R.R., Huslter, K., & Mundy, P., J. (1996). The Impact of Man on Raptors in
Zimbabwe. In D.Bird., D. Varland & J. Negro, (Ed.) Raptors in Human
Landscapes (pp. 215-231). London, England: Academic Press.
58
Hinam, H. L., & Clair, C. C. (2008). High levels of habitat loss and fragmentation limit
reproductive success by reducing home range size and provisioning rates of Northern
saw-whet owls. Biological Conservation, 141(2), 524-535.
Horton, S. (1996). Spotted Owls in Managed Forests of Western Oregon and
Washington. In D.Bird., D. Varland & J. Negro, (Ed.) Raptors in Human
Landscapes (pp. 215-231). London, England: Academic Press.
Hovick, T, J., Elmore, R.D., Dahlgren, D. K., Fuhlendorf, S. D., & Engle, D., M. (2014).
Evidence of negative effects of anthropogenic structures on wildlife: a review of
grouse survival and behaviour. British Ecological Society, 51(6): 1680-1689.
How Kernel Density works. (n.d.) ESRI. Retrieved from https://pro.arcgis.com/en/proapp/latest/tool-reference/spatial-analyst/how-kernel-density-works.htm
Huijser, M.P., Galarus, D.E., & Kociolek, A.V., (2010). Current and developing
technologies in highway-wildlife mitigation. In J.P. Beckmann, A.P. Clevenger,
M.P. Huijser & J.A. Hitly, (Ed.) Safe Passages: Highway, Wildlife, and Habitat
Connectivity (pp. 309-321). Washington, DC: Island Press.
Huijser, M.P., Mosler-Berger, C., Olsson, M., & Strein, M. (2015). Wildlife warning
signs and detection systems aimed at reducing wildlife-vehicle collisions. In R.
van der Ree, D.J.,Smith, and C. Grilo, (Ed.) Handbook of Road Ecology (pp. 198212). John Wiley & Sons, Ltd.
Husby, M. (2016). Factors affecting road mortality in birds. Ornis Fennica, 93, 212-224.
59
Husby M. (2017). Traffic influence on roadside bird abundance and behaviour. Acta
Ornithol, 52, 93–103.
Jacobson, S.L .(2005). Mitigation Measures for Highway-caused Impacts on Birds,
USDA Forest Service Gen. Tech. Rep. PSW-GTR-191.1043-1050.
Jackson Hole Wildlife Foundation. Give Wildlife a Brake.
https://jhwildlife.org/our-work/give-wildlife-a-brake/
Jones, D.N., & Bond, A.R.F. (2010). Road barrier effect on small birds removed by
vegetated overpass in South East Queensland. Ecological Management, 11(1), 65-67.
Kinyon, J. (2017, September 13). Flattened Forms: Using Survey 123 for ArcGIS to
eliminate paper data sheets and collect roadkill information. ESRI.
https://community.esri.com/t5/arcgis-survey123-blog/flattened-forms/bap/897395?commentID=51211
Kociolek, A. V., Clevenger, A. P., St. Clair., & Proppe, D. S. (2011). Effects of road
networks on bird populations. Conservation Biology, 25( 2), 241-249.
Kociolek, A., Grilo, C., & Jacobson, S. (2015). Flight doesn’t solve everything:mitigation
of road impacts on birds. In R. van der Ree, D.J., Smith, and C. Grilo,
(Ed.)Handbook of Road Ecology (pp. 281-289). John Wiley & Sons, Ltd.
Kovacs, A., Mammen, U, C.C., & Wernham, C.V. (2008). European Monitoring for
Raptors and Owls: State of Art and Future Needs. Ambio, 37 (6): 408-412.
60
Lambertucci, S.A., Speziale, K. L., Rogers, T.E., & Morales, J.M. (2009). How do roads
affect the habitat use of an assemblage of scavenging raptors? Biodivers Conserv, 18,
2063-2074.
Lee, T., Quinn, M., & Duke, D. (2010). A local community monitors wildlife along a
major transportation corridor. In J.P. Beckmann, A.P. Clevenger, M.P. Huijser &
J.A. Hitly, (Ed.) Safe Passages: Highway, Wildlife, and Habitat Connectivity (pp.
277-292). Washington, DC: Island Press.
Lee, T. & Rondeau, K. (2018). Collision Count: Improving human and wildlife safety on
Highway 3. Miistakis Institute.
https://www.rockies.ca/files/reports/CollisionCount_Reporting2018_Final.pdf
Lee, T. S., Rondeau, K., Schaufele, R., Clevenger, A.P., & Duke, D. (2021). Developing
a correction factor to apply to animal vehicle collision data for improved road
mitigation measures. Wildlife Research, A-J.
Lindenmayer D.B. & Fischer, J. 2006. Habitat Fragmentation and Landscape Change.
Island Press.
Loos, G. & Kerlinger, P. 1993. Road Mortality of Saw-Whet and Screech-Owls on the
Cape May Peninsula. Journal of Raptor Research 27(4): 210-213.
Loss, S. R., Will, T. & Marra, P. P. (2014). Estimation of Bird-Vehicle Collision
Mortality on U.S. Roads. The Journal of Wildlife Management, 78 (5), 763-771.
McClure, C. J.W., Westrip, J. R.S., Johnson, J. A., Schulwitz, S. E., Virani, M. Z.,
61
Davies, R., Symes, A., Wheatley, H., Thorstrom, R., Amar, A., Buij, R., Jones,
V.R., Williams, N. P., Buechley, E. R., &Butchart, S. H. (2018). State of the
world’s raptors: Distributions, threats, and conservation recommendations.
Biological Conservation, 227, 390-402.
McClure, C.J.W., Schulwitz, S.E., Anderson, D.L., Robinson, B.W., Mojica, E.K.,
Therrien, J., Oleyor, M.D., Johnson, J. (2019). Commentary: Defining Raptors
and Birds of Prey. Journal of Raptor Research, 53 (4): 419-430.
Meunier, F. D., Verheyden, C., & Jouventin, P. (2000). Use of roadsides by diurnal
raptors in agricultural landscapes. Biological Conservation, 92, 291-298.
Newton, I. (1979). Population Ecology of Raptors. Buteo Books.
Nguyen, T., Saleh, M-K., Kyaw, M., Trujillo, G., Bejarano, M., Tapia, K., Waetjen, D.,
& Shilling, F. (2020, June 24). Special Report 4: Impact of COVID-19 Mitigation
on Wildlife-VehicleConflict (Rep.).
https://roadecology.ucdavis.edu/files/content/projects/COVID_CHIPs_Impacts_w
ildlife.pdf
O’Brien, E. (2006). Habitat fragmentation due to transport infrastructure: Practical
considerations. In J. Davenport and J.L. Davenport (Ed.) The Ecology of Transportation:
Managing Mobility for the Environment (pp.275-325). Netherlands: Springer.
Orlowski, G. & Siembieda, J. (2005). Skeletal injuries of passerines caused by road
traffic. Acta Ornithologica, 40(1), 15-19.
62
Patten, M.A., Wolfe, D.H., Sochat, E, & Sherrod, S.K. (2005). Habitat fragmentation,
rapid evolution, and population persistence. Evolutionary Ecology Research, 7,
235-249.
Paul, K. J. S. Auditing a Monitoring Program: Can Citizen Science Document Wildlife
Activity Along Highways? Master’s Thesis, The University of Montana. 1-61.
Pell, S., & Jones, D. (2015). Are wildlife overpasses of conservation value for birds? A
study in Australian sub-tropical forest, with wider implications. Biological
Conservation, 184, 300-309.
Petty, S. J. (1996). Adaptations of Raptors to Man-Made Spruce Forests in the Uplands
of Britain. In D.Bird., D. Varland & J. Negro, (Ed.) Raptors in Human
Landscapes (pp. 201-214). London, England: Academic Press.
Pons, P. (2000). Height of the road embankment affects probability of traffic collision by
birds. Bird Study, 47, 122-125.
Ramsden, D. J. (2007). Barn Owls and Major Roads. The Barn Owl Trust.
https://www.barnowltrust.org.uk/wp-content/uploads/Barn_Owls_and_Major_Roads.pdf
Santos, S.M., Mira, A., Salgueiro, P.A., Costa, P., Medians, D., & Beja. P. (2016). Avian
trait-mediated vulnerability to road traffic collisions. Biological Conservation, 200, 122130.
Seiler, A. & Helldin, J-O. (2006). Mortality in Wildlife due to Transportation. In J.
Davenport and J.L. Davenport (Ed.) The Ecology of Transportation: Managing
Mobility for the Environment (pp.165-189). Netherlands: Springer.
63
Sevigny, J., Summer, A, Kalisz, G., & McAllister, K. (2021). Idenitification of Elkvehicle incident hotspots on state route 20 in Washington State. Landscape
Ecology, 1-14.
Shilling, F., Perkins, S.E., & Collinson, W. (2015). Wildlife/Roadkill Observation and
Reporting Systems. In R. van der Ree, D.J., Smith, and C. Grilo, (Ed.) Handbook of Road
Ecology (pp. 492-501). John Wiley & Sons, Ltd.
Spellerberg, I.F. (1998). Ecological Effects of Roads and Traffic: A Literature Review.
Global Ecology and Biogeography Letters, 7(5), 317-333.
Smith, D. J. & van der Ree, R. (2015). Field Methods to Evaluate the Impacts of Roads
on Wildlife. In R. van der Ree, D.J., Smith, and C. Grilo, (Ed.) Handbook of Road
Ecology (pp. 82-95). John Wiley & Sons, Ltd.
Stewart, B. (2019) Assessing the permeability of large underpasses and viaducts on
Interstate 5 in Southwest Washington State for local wildlife, with an emphasis on
ungulates. Master’s Thesis, Evergreen State College. 1-271.
Taylor, J. (2021, April). Protecting Scavenging Eagles from Vehicle Strikes. PowerPoint
presentation, HawkWatch International.
Tigas, L.A., Van Vuren, D.H. & Sauvajot, R.M. (2002). Behavioral responses of bobcats
and coyotes to habitat fragmentation and corridors in an urban environment.
Biological Conservation, 108, 299-306.
64
van der Ree, R., Jaeger, J. A.G., van der Grift, E. A., & Clevenger, A.P. 2011. Effects of
Roads and Traffic on Wildlife Populations and Landscape Function: Road
Ecology is Moving toward Larger Scales. Ecology and Society, 16(1): 48.
Washington Department of Fish and Wildlife. (2009). Landscape Planning for
Washington’s Wildlife: Managing for Biodiversity in Developing Areas. WDFW.
Washington Department of Transportation. (2016). 2016 Annual Traffic Report.
WSDOT.https://wsdot.wa.gov/mapsdata/travel/pdf/Annual_Traffic_Report_2016.pdf
Washington Department of Transportation. (2021). Reducing the risk of wildlife
collisions. WSDOT: https://wsdot.wa.gov/environment/protecting/wildlife-collisions
Why Raptors? (n.d.). Retrieved from https://hawkwatch.org/learn/why-raptors.
Wilcove D. S, McLellan C. H, Dobson A. P. (1986). Habitat fragmentation in the
temperate zone. Conservation Biology, 237–256.
World Population Review. (2021). Washington Population 2021. Retrieved from
https://worldpopulationreview.com/states/washington-population
65
Appendices
Appendix A: Washington State Highway Amount of Vehicle Miles Traveled per year 19902016 The amount of vehicle miles traveled (AVMT) in Washington state from 1990-2016. The
expanded data is separated by rural and urban miles and shows the percent change from year to
year. Table is taken from 2016 Annual Traffic Report, Washington Department of
Transportation, pp.55.
66
Appendix B: Complete list of Raptor Species included in Arc Survey123
Common Name
Scientific Name
Turkey Vulture
Osprey
White-tailed Kite
Golden Eagle
Northern Harrier
Sharp-shinned Hawk
Cooper's Hawk
Northern Goshawk
Bald Eagle
Red-shouldered Hawk
Swainson's Hawk
Red-tailed Hawk
Rough-legged Hawk
Ferruginous Hawk
Barn Owl
Flammulated Owl
Western Screech-Owl
Great Horned Owl
Snowy Owl
Northern Hawk Owl
Northern Pygmy-Owl
Burrowing Owl
Northern Spotted Owl
Barred Owl
Great Gray Owl
Long-eared Owl
Short-eared Owl
Boreal Owl
Northern Saw-whet Owl
American Kestrel
Merlin
Peregrine Falcon
Prairie Falcon
Cathartes aura
Pandion haliaetus
Elanus leucurus
Aquila chrysaetos
Circus cyaneus
Accipiter striatus
Accipiter cooperii
Accipiter gentilis
Haliaeetus leucocephalus
Buteo lineatus
Buteo swainsoni
Buteo jamaicensis
Buteo lagopus
Buteo regalis
Tyto alba
Otus flammeolus
Megascops kennicottii
Bubo virginianus
Bubo scandiacus
Surnia ulula
Glaucidium gnoma
Athene cunicularia
Strix occidentalis
Strix varia
Strix nebulosa
Asio otus
Asio flammeus
Aegolius funereus
Aegolius acadicus
Falco sparverius
Falco columbarius
Falco peregrinus
Falco mexicanus
Group
New World
Vulture
Osprey
Kite
Eagle
Harrier
Hawk
Hawk
Hawk
Eagle
Hawk
Hawk
Hawk
Hawk
Hawk
Owl
Owl
Owl
Owl
Owl
Owl
Owl
Owl
Owl
Owl
Owl
Owl
Owl
Owl
Owl
Falcon
Falcon
Falcon
Falcon
67
