Roads are one of the most recognizable artifacts of human habitation and development. They cross various landscapes and expand with the increasing density of human settlements. While these structures are vital in facilitating vehicular transportation, they can also have unintended negative consequences for natural ecosystems, such as road-related wildlife mortality. To examine the negative impacts of roads on wildlife at a local level, I quantified the frequency and patterns of road mortality across a yearlong study period in a study area within Montgomery and Giles Counties in southwest Virginia. I also examined relationships between roadkill patterns and several land and road variables, hoping to close a current literature gap by focusing specifically on small and meso-mammals weighing less than 15kg. I recorded the location and species of carcasses found along predetermined routes, including roads with variable speed limits, road infrastructure, and adjacent land cover. I used government-provided data sets, satellite imagery, and field verification to characterize our study routes. I recorded 294 carcasses and identified 221, of which 84% were small or meso-mammals, including 11 taxa. The most frequent roadkill taxa were tree squirrels (Sciurus sp.), opossums (Didelphis virginiana), striped skunks (Mephitus mephitis), and raccoons (Procyon lotor). I found three roadkill hotspots, all along a major highway, and a high frequency of roadkill in spring, autumn, and early winter. High posted speed limits, the presence of artificial road lighting, and high proportions of forest and early successional growth cover (often present in pastures) were associated with an increased frequency of wildlife-vehicle collisions. / Roads are one of the most recognizable artifacts of human habitation and development. They cross various landscapes and expand with the increasing density of human settlements. While these structures are vital in facilitating automobile transportation, they can also have unintended negative consequences for natural ecosystems, such as roadkill. To examine the negative impacts of roads on local wildlife, I studied the frequency, patterns, and causes of road mortality across a year-long study period. I also examined relationships between frequent roadkill hotspots and several land and road variables in a Montgomery and Giles Counties study area in southwest Virginia. My project focused on small and medium-sized mammals because they are rarely considered in other roadkill studies. I collected field data on the location and species of carcasses observed on surveys along predetermined routes, including roads with variable speed limits, road infrastructure, and adjacent land cover. I used government-provided data sets, imagery, and field verification to characterize study routes. I recorded 294 carcasses and identified 221, 84% of which were small and medium-sized mammals. Roadkill frequency was highest in spring, autumn, and late winter. I found that high posted speed limits, the presence of artificial lighting, high forest cover, and high early successional growth cover (such as shrubbery and small trees in pastures) were associated with a higher frequency of wildlife-vehicle collisions.

There is an expansive network of roads in Algonquin Provincial Park (APP) to facilitate forestry resource extraction. This leaves a research need for examining how the logging road network in APP affects the large mammals, and what local-level and landscape-level variables influence that use. Local-level data was collected directly at observation points, and landscape-level data was produced from ArcGIS for 40km2, 80km2, and 130km2 buffer areas. The objective of my study was to look at the use of primary and branch logging roads by five large mammal species in APP, and determine if landscape-level variables had an influence on the level of movement and utilization. The five species included moose, white-tailed deer, American black bear, eastern wolf and coyote. My null hypothesis (H0) states that there will be no difference in large mammal use between the primary and branch logging roads within APP and that local- and landscape-level variables will have no influence on them; my alternative hypothesis (H1) states that there will be less large mammal activity on the primary logging roads, more large mammal movement on the branch logging roads and local- and landscape-level variables will influence this use. Tracking was done by vehicle on six transects across the park for three repeated surveys where species identification and local-level variables were recorded. Landscape-level variables were acquired through GIS analysis in the lab. Based on the results from the local-level data, branch and primary logging road use differed in composition, though no significance was found between the use by large mammals for these two types of road. Through generalized linear models, specific combinations of landscape-level variables did influence large mammal movement on the primary and branch logging roads within three habitat range scales (130km2, 80km2, and 40km2). The most significance was seen at the buffer of 40km2 on the branch logging roads, with the variables road density (p < 0.01), percent forest cover (p = 0.04) and topographic ruggedness (p < 0.01) all having a strong impact on large mammal movement. The only significant findings for primary logging roads were also at the 40km2 scale with percent forest cover (p = 0.03) and percent water cover (p = 0.02) having an impact on large mammal movement. Overall, the landscape variables had greater influence on branch logging roads that may be explained by the quality of the surrounding habitat, as well as greater influence at smaller buffer scales. Further research and monitoring of the large mammals in APP is recommended to expand on this preliminary study. Greater understanding of the local- and landscape-level variables at differing habitat ranges will assist in understanding these large mammal movements and provide data to base logging road management on. As large mammals are wide-ranging species, my study informs APP that their logging road network does not seem to hinder the movements of this group of animals. Overall, the large mammals in APP did not have any significant difference in their use of primary and branch logging roads of APP. Further research has the potential to give greater understanding of the impacts of the logging road network on the five large mammal species studied in APP. There is also the potential for useful management strategies to emerge for large mammals in this park, and how to incorporate human activities within their habitat while maintaining sustainable populations.

As land-use change threatens wildlife viability, the understanding of how animals move through spatially fragmented landscapes has important implications for the long-term persistence and management of species. This research compares moose movement and space-use patterns between Algonquin Provincial Park (20 moose) and Wildlife Management Unit 49 (17 moose) in southern Ontario from 2006 to 2008. Moose were found to be using areas in the home range that are not normally used more often in the park relative to the management unit, and that there are signs, although not significant, that the use of land covers between the two locations may be diverging. These differences are likely driven by the differences in predator and road densities between the two locations. I interpret these findings to suggest that Algonquin Provincial Park is protecting moose from the impacts of roads, but that moose can survive in the road-fragmented landscapes if there is enough habitat available.

As land-use change threatens wildlife viability, the understanding of how animals move through spatially fragmented landscapes has important implications for the long-term persistence and management of species. This research compares moose movement and space-use patterns between Algonquin Provincial Park (20 moose) and Wildlife Management Unit 49 (17 moose) in southern Ontario from 2006 to 2008. Moose were found to be using areas in the home range that are not normally used more often in the park relative to the management unit, and that there are signs, although not significant, that the use of land covers between the two locations may be diverging. These differences are likely driven by the differences in predator and road densities between the two locations. I interpret these findings to suggest that Algonquin Provincial Park is protecting moose from the impacts of roads, but that moose can survive in the road-fragmented landscapes if there is enough habitat available.

ABSTRACT The greatest driver of the current global biodiversity crisis is habitat loss. Roads are a major contributor to habitat loss because they destroy and fragment habitat, in addition to causing direct mortality. Animals may respond to roads either by avoiding them, thus leading to population isolation, or by attempting to cross them, thus potentially leading to increased mortality and, if so, also to population isolation. I studied the impact of road density on abundance of two northern snake species: the redbelly snake (Storeria occipitomaculata) and the garter snake (Thamnophis sirtalis). I hypothesized that roads are detrimental to snake populations due to road avoidance and road mortality. Therefore, I predicted that snakes should be less abundant in sites with higher road density in their surroundings. I deployed cover boards at 28 old field sites along a gradient of road density in 2020 and in 2021. I visited sites weekly and counted the number of individuals of both species. I captured fewer garter snakes at sites surrounded by more roads, and fewer redbelly snakes at sites enclosed by more roads. The effect of roads on number of snakes is modest, but could be indicative of decreasing population size, which could in turn lead to loss of ecological function. RÉSUMÉ Le plus grand moteur de la crise mondiale actuelle de la biodiversité est la perte d'habitat. Les routes contribuent grandement à la perte d'habitat parce qu'elles détruisent et fragmentent l'habitat, en plus de causer de la mortalité directe. Les animaux peuvent réagir aux routes soit en les évitant, entraînant ainsi l'isolement des populations, soit en tentant de les traverser, entraînant ainsi potentiellement une mortalité accrue et également l'isolement des populations. J'ai étudié l'impact de la densité des routes sur l'abondance de deux espèces de couleuvres nordiques : la couleuvre à ventre rouge (Storeria occipitomaculata) et la couleuvre rayée (Thamnophis sirtalis). J'ai émis l'hypothèse que les routes sont néfastes pour les populations de serpents en raison de l'évitement des routes et de la mortalité routière. Par conséquent, j'ai prédit que les couleuvres devraient être moins abondantes dans les sites avec une densité routière plus élevée dans leurs environs. J'ai déployé des plaques abris sur 28 sites de champs en friche le long d'un gradient de densité de routes en 2020 et en 2021. J'ai visité les sites chaque semaine et compté le nombre d'individus des deux espèces. J'ai capturé moins de couleuvres rayées et moins de couleuvres à ventre rouge aux sites entourés de plus de routes. L'effet des routes sur le nombre de couleuvres est modeste, mais pourrait indiquer une diminution de la taille de la population, ce qui pourrait à son tour entraîner une perte de fonction écologique.

Road expansion and increased traffic likely exacerbates barriers to amphibian migration and dispersal. Within British Columbia’s south Okanagan valley there is particular concern that the COSEWIC-listed blotched tiger salamander (Ambystoma mavortium melanostictum) and Great Basin spadefoot (Spea intermontana) are vulnerable to road effects in their annual movements from upland overwintering habitat to lowland breeding areas. My study utilizes a before after control impact approach to assess amphibian movement and population threats across this highway-bisected landscape. Throughout the spring and summer of 2010-2012, fifty two kilometers of roadways (31 km of highway, 21 km of paved backroad) were repeatedly surveyed from the Canada-USA border to north of Oliver, BC; surveys were carried out utilising vehicles and on foot. Along Highway 97, a three kilometer four-lane highway expansion project was constructed through 2010 and open to traffic use in 2011. Adjacent to a floodplain, survey effort was focused throughout this transect for informed roadkill mitigation structure placement and ongoing ecopassage effectiveness monitoring. Automated camera trap monitoring of culverts within highly concentrated amphibian road hotspots during spring and summer 2011 (three culverts) and 2012 (two culverts) resulted in over eight hundred amphibian culvert events observed. Two sample Wilcoxon tests revealed differences between years in amphibian occurrence between 2010 and 2012 (W = 4679.5, p= 0.02), and mortalities among transect areas, with the largest differences between years within the Osoyoos passing lanes transect. Amphibian mortalities within the passing lanes transect were significantly reduced with the implementation of mitigation structures (x̅2010= 13.2 ± 32.5, x̅2011= 4.7 ± 12.8, x̅2012= 2.3 ± 7.3; 2010 vs. 2012: W= 1535.5, p< 0.001). Roadkill mitigation structures proved effective in observed amphibian occurrence of the entire passing lanes stretch as well as at distances 100 m and 200 m from observed culverts. Double fenced areas resulted in a 94% reduction in amphibian road occurrence. Five species of amphibians were observed over the three survey years (4051 road incidences over 657 survey hours): Pacific chorus frog (Pseudacris regilla), Western toad (Anaxyrus boreas), long-toed salamander (Ambystoma macrodactylum) plus blotched tiger salamander and Great Basin spadefoot. This study aims to provide a better understanding of amphibian hotspots on roadways and ecopassage use within the south Okanagan. It may act as a catalyst to further wildlife-vehicle interaction studies with improved mitigation solutions for amphibian roadway fatalities.

There is growing global concern over the influence of road development on the conservation of biodiversity and on the functioning of ecosystems. Published reviews in the field of road ecology have identified that most research has examined the effects of roads linearly and have advocated for research at landscape scales. Among the many effects roads have, one of the most significant is the loss of animal life resulting from collisions with vehicles. Despite this, little is known of what toll this has on animal populations and how these impacts vary with scale. This stems from the perception that impacts are localised and that animals killed are typically considered common, and therefore not of great conservation concern. This thesis challenges this notion by showing that the impacts of fatalities can affect populations at landscape extents and that commonness is not a barrier to localised extinction risk. To achieve this I focus on the common wombat; an example of a common species for which road impacts have never been previously examined. Chapter 1 provides an overview of the importance of scale in quantifying road impacts and the debate surrounding common species persistence in road environments. Chapter 2 assesses habitat use of wombats in a road environment at a local scale. Results suggest that wombats select for roadside habitat and as a result populations could be under threat. Chapter 3 is a predictive model of wombat road fatalities which demonstrates the importance of incorporating habitat use in predictive fatality modelling. Through use of a spatially explicit population viability analysis, Chapter 4 demonstrates that roads, in conjunction with other threats can affect the persistence of a common species at a local scale. Chapter 5 is a landscape extent assessment of wombat habitat use, finding that increased effort should be employed in evaluating how reserves confer resilience to species from the impacts of roads and that habitat quality can dictate road-based fatality rates. Chapter 6 summarises the research presented in the thesis and suggests direction for future work, particularly the importance of evaluating the interplay between susceptibility and abundance on species vulnerability in road environments.

Roads pose two central problems for wildlife: wildlife-vehicle collisions (WVCs) and habitat fragmentation. Wildlife exclusion fencing can reduce WVCs but can exacerbate fragmentation. In Chapter 1, I summarize the relevant studies addressing these two problems, with a focus on large mammals in North America. Chapters 2 and 3 summarize field assessments of technologies to reduce WVCs and maintain connectivity, specifically jumpout ramps and underpasses, along Highway 101 near San Luis Obispo, CA. In a fenced highway, some animals inevitably breach the fence and become trapped, which increases the risk of a wildlife-vehicle collision. Earthen escape ramps, or “jumpouts”, can allow the trapped animal to escape the highway corridor. Few studies have quantified wildlife use of jumpouts, and none for >2 years. We used wildlife cameras to quantify wildlife use of 4 jumpouts from 2012-2017. Mule deer were 88% percent of our detections and jumped out 20% of the time. After accounting for pseudoreplication, 33% of the events were independent events, and 2 groups of deer accounted for 41% of all detections at the top of the jumpout. Female deer were 86% of the detections and were much more likely than males to return to the jumpout multiple times. This is the first study to document use of jumpouts for more than 3 years, the first to account for pseudoreplication, and the first to quantify differences in jumpout use between male and female mule deer. We recommend a jumpout height between 1.75m-2m for mule deer to increase the jumpout success rate. Chapter 3 addresses factors that may affect the use of undercrossings by mule deer and other wildlife. Wildlife crossings combined with wildlife exclusion fencing have been shown to be the most effective method to reduce wildlife-vehicle collisions while maintaining ecological connectivity. Although several studies have quantified wildlife use of undercrossings, very few have exceeded 24 months, and the factors affecting carnivores use of the undercrossings remain unclear. We quantified mule deer, black bear, mountain lion, and bobcat use of 11 undercrossings along Highway 101 near San Luis Obispo, California from 2012-2017. We constructed zero-inflated Poisson general linear models on the monthly activity of our focal species using underpass dimensionality, distance to cover, substrate, human activity, and location relative to the wildlife exclusion fence as predictor variables. We accounted for temporal variation, as well as spatial variation by quantifying the landscape resistance near each undercrossing. We found that deer almost exclusively used the larger underpasses whereas the carnivores were considerably less selective. Bears used undercrossings more that were within the wildlife exclusion fence, whereas mountain lion activity was higher outside the wildlife exclusion fence. Bobcat activity was highest and most widespread, and was negatively associated with distance to cover. Regional connectivity is most important for bear and mountain lion, and the surrounding habitat may be the most important predictor for their use of undercrossings. We recommend placing GPS collars on our focal species to more clearly document fine-scale habitat selection near the highway.

As rodovias podem representar um importante fator na fragmentação de habitat para espécies silvestres adaptadas a habitats de alta complexidade estrutural, como as florestas tropicais. As estradas reduzem a conectividade da paisagem e a capacidade da população regional em habitar todas as áreas adequadas e estes efeitos são mais significativos nas espécies que evitam a estrada, que são, muitas vezes, espécies de interior de florestas. A magnitude dos efeitos de barreira dependerá do comportamento e mobilidade destas espécies. Quando as estradas representam ralos (sink) ou barreiras para as populações, devido, respectivamente, aos atropelamentos ou à repulsa, medidas mitigadoras são indicadas para aumentar a conectividade entre as manchas de habitat separadas por essas estradas. A qualidade do habitat é um fator que deve ser considerado, mesmo com baixas frequências de atropelamentos nesses locais. O objetivo desse estudo foi propor dois métodos de seleção de áreas prioritárias para implantação de medidas mitigadoras dos efeitos das estradas sobre espécies de vertebrados florestais: Seleção Hierárquica Multivariada e Seleção Bivariada. A área de estudo foi o bioma Mata Atlântica, sendo recortado em paisagens hexagonais em três escalas diferentes (10.000, 1.000 e 100 km), usando a extensão Repeating Shapes no programa ArcGIS 9.3. Em cada hexágono foram calculados: área de floresta e de Unidade de Conservação, densidade de estradas e de hidrografia. Apenas os hexágonos cobertos por no mínimo 45% pela Mata Atlântica, com mais de 50% de cobertura florestal e mais de 1% de Unidades de Conservação foram incluídos nas análises. Após esta seleção, no método Seleção Hierárquica Multivariada, foi feita uma análise de componentes principais (PCA) com as quatro variáveis medidas, para cada escala separadamente. Os hexágonos foram então ordenados segundo o posicionamento deles no 1 Eixo da PCA de forma hierárquica e da maior para menor escala de hexágonos. Para área de estudo o método de Seleção Bivariada foi construído um gráfico de pontos, para cada escala de hexágono, com as variáveis cobertura florestal e rios. Foram selecionados os hexágonos que estavam localizados no quadrante do gráfico que representasse maior densidade de rios e maior porcentagem de cobertura florestal. Posteriormente foi feita uma simulação para avaliar se os métodos eram capazes de recuperar escores tão alto quanto a ordenação seguindo apenas o posicionamento dos hexágonos no Eixo 1 da PCA, sem uma análise hierárquica. O método de Seleção Hierárquica Multivariada foi mais eficiente para escolha de áreas prioritárias do que a Seleção Bivariada tanto para a escala intermediária (1.000 km) quanto para a menor escala (100 km). Os cinco hexágonos de 100 km mais prioritários estão localizados em São Paulo e Paraná, abrangendo quatro UCs (PARES de Jacupiranga, APA de Guaraqueçaba, APA Cananéia- Jacuípe e PARES da Ilha do Cardoso). Devido à simplicidade e fácil aplicabilidade do método, acredita-se que este pode ser uma opção interessante para escolha de áreas prioritárias para implantação de medidas mitigadoras dos efeitos de estradas / Roads may represent an important factor in fragmentation of habitat for wild species adapted to habitats of high structural complexity as rainforests. Roads reduce landscape connectivity and the capacity of the regional population live in all suitable areas and these effects are most significant in species that avoid the road, which are often species dependent on forests. The magnitude of the barrier effects depend on the behavior and mobility of these species. When roads represent "sinkers" or barriers for populations due, respectively, trampling or avoidance, mitigation measures are indicated to increase connectivity between habitat patches separated by these roads. The habitat quality is a factor that must be considered, even at low frequencies of road kill there. The goal of this study was to propose two methods of selecting priority areas for implementation of mitigation measures of roads effects on forest vertebrate species: Multivariate Hierarchical Selection and Bivariate Selection. The study area was the Atlantic Forest biome, being cut into hexagonal landscapes at three different scales (10,000, 1,000 and 100 km), using the extension "Repeating Shapes" program in ArcGIS 9.3. In each hexagon was calculated: area of forest, protected areas, density of roads and hydrograph. Only hexagons covered by at least 45% by the Atlantic Forest, with over 50% forest cover and more than 1% of protected areas were included in the analyzes. After this using the Multivariate Hierarchical selection method, we performed a principal component analysis (PCA) with the four variables measured for each scale separately. The hexagons were then ordered according to their position in the first axis of PCA, using a hierarchical manner and the smaller scale of hexagons. For the Bivariate selection method we did a dot plot for each scale hexagon, with the variables forest cover and rivers. We selected hexagons that were located in the quadrant of the graph that represented higher density of rivers and larger percentage of forest cover. After that, we made a simulation to evaluate if the two methods were able to retrieve scores as high as the ordination following only the positioning of the hexagons on Axis 1 of the PCA, without a hierarchical analysis. For the study site, the Multivariate Hierarchical selection was more efficient for choosing priority areas than Bivariate Selection for both the intermediate scale (1000 km) and to a lesser extent (100 km). The five hexagons of 100 km of highest priorities are located in São Paulo and Paraná, covering four protected areas (PARES of Jacupiranga, Guaraqueçaba APA, and APA Cananéia-Jacuípe and PARES Ilha do Cardoso). Due to simplicity and easy applicability of the method, we believe this may be an interesting option for selecting priority areas for implementation of mitigation measures of the effects of roads

Atropelamentos de animais silvestres são a principal causa de mortalidade de origem antrópica de vertebrados terrestres. Além da mortalidade direta, as populações animais também são fragmentadas e isoladas por rodovias, que podem atuar como filtro ou barreira ao movimento da fauna. A indicação e implementação de medidas mitigadoras têm sido uma estratégia cada vez importante, ampliando a necessidade de desenvolver e qualificar métodos para avaliar os impactos e indicar áreas prioritárias. Essa tese de doutorado foi concebida com a preocupação de investigar certos temas relacionados à mortalidade e fragmentação por rodovias. No primeiro capítulo, discuto como a qualificação da pesquisa e do licenciamento podem colaborar com este cenário. No segundo capítulo, apresento uma revisão de diferentes métodos de análise espacial utilizados para testar se existe a presença de agregações de atropelamento e para localizar onde estão estas agregações. No terceiro capítulo, apresento os resultados de um modelo de simulação baseado em indivíduos, que mostra que a localização dos hotspots muda ao longo do tempo em função da diminuição das populações próximas a trechos de rodovias com maior letalidade, o que torna a mortalidade per capita um melhor indicador da necessidade de mitigação. No último capítulo avaliei o efeito da rede de rodovias na fragmentação de habitat nos campos sulinos do Rio Grande do Sul, e demonstro que considerar o efeito da rede de rodovias como uma barreira aos movimentos da fauna modifica de forma severa a percepção que temos sobre o status de conservação dos campos. Esta tese pode ter dois tipos principais de implicações: a aplicação direta dos resultados aqui apresentados nas avaliações dos impactos de rodovias e planejamento da mitigação, e a influência em novos rumos de pesquisa na ecologia de rodovias. / Roads are responsible for a series of impacts to ecosystems, and some authors point out that road-kills are the main cause of terrestrial vertebrate mortality from anthropogenic causes. Besides direct mortality, wildlife populations are also fragmented and isolated by roads, as they can act as barriers or filters to wildlife movement. Implementing mitigation measures had become an important conservation strategy, but the need to prioritize areas brings the urgency to develop and qualify methods to assess road impacts and indicate priority areas. This doctorate thesis was developed with the concern of investigating subjects related to wildlife mortality and fragmentation by roads. In the first chapter I discuss how qualifying research and environmental licensing may contribute in this scenario. In the second chapter, I present a review of different methods of spatial analysis that have been used to test the presence of clustering on road-kill data and to identify road-kill hotpots. In the third chapter, I present the results of simulations of an individual-based model that shows that the location of road-kill hotspots change in time due to population depression near high-risk road segments; making per capita mortality a better indicator of the need for mitigation. In the last chapter, I evaluated the effect of the road network on habitat fragmentation of South Brazilian grasslands in Rio Grande do Sul State, and I show that considering the road network as a barrier changes severely our perception about grassland conservation status. This thesis may have two types of implications: the direct applications of the results presented here in environmental impact assessment of roads and in mitigation planning, or the influence on new paths to study road effects on wildlife.