Roadkill Research Articles

Anthropogenic mortality threatens the survival of Canarian houbara bustards

Anthropogenic mortality is a major cause of global mortality in terrestrial vertebrates. Quantifying its impact on the dynamics of threatened species is essential to improve their conservation. We investigated cause-specific mortality in Canarian houbara bustards (Chlamydotis undulata fuertaventurae), an endangered subspecies endemic to the Canary Islands. We monitored 51 individuals tagged with solar-powered GSM/GPRS loggers for an average of 3.15 years, and recorded 7 casualties at aerial lines (13.73% of the sample; 5 at power lines, 2 at telephone lines), 1 (1.96%) at a wire fence, 4 road kills (7.84%) and 1 case of predation by cat (1.96%). Cox proportional hazards models showed that anthropogenic and natural annual mortality rates were similar (respectively, 6.20% and 6.36% of the individuals). We estimate that 33–35 houbaras die each year in the Canary Islands due to anthropogenic causes. Population viability models using these data and juvenile productivity values obtained over seven years predicted the extinction of the species in 50 years. Eliminating anthropogenic mortality, the population could be recovered, but would still require management actions to improve habitat quality. Conservation measures to reduce anthropogenic mortality due to power line fatalities, roadkills and predation by cats, as well as to increase productivity, are urgently needed, particularly on Fuerteventura, where houbaras are on the brink of extinction.

Evaluating YOLO architectures for detecting road killed endangered Brazilian animals

Wildlife roadkill is a recurring, dangerous problem that affects both humans and animals and has received increasing attention from environmentalists worldwide. Addressing this problem is difficult due to the high investments required in road infrastructure to effectively reduce wildlife vehicle collisions. Despite recent applications of machine learning techniques in low-cost and economically viable detection systems, e.g., for alerting drivers about the presence of animals and collecting statistics on endangered animal species, the success and wide adoption of these systems depend heavily on the availability of data for system training. The lack of training data negatively impacts the feature extraction of machine learning models, which is crucial for successful animal detection and classification. In this paper, we evaluate the performance of several state-of-the-art object detection models on limited data for model training. The selected models are based on the YOLO architecture, which is well-suited for and commonly used in real-time object detection. These include the YoloV4, Scaled-YoloV4, YoloV5, YoloR, YoloX, and YoloV7 models. We focus on Brazilian endangered animal species and use the BRA-Dataset for model training. We also assess the effectiveness of data augmentation and transfer learning techniques in our evaluation. The models are compared using summary metrics such as precision, recall, mAP, and FPS and are qualitatively analyzed considering classic computer vision problems. The results show that the architecture with the best results against false negatives is Scaled-YoloV4, while the best FPS detection score is the nano version of YoloV5.

Leveraging LiDAR-Based Simulations to Quantify the Complexity of the Static Environment for Autonomous Vehicles in Rural Settings.

This paper uses virtual simulations to examine the interaction between autonomous vehicles (AVs) and their surrounding environment. A framework was developed to estimate the environment's complexity by calculating the real-time data processing requirements for AVs to navigate effectively. The VISTA simulator was used to synthesize viewpoints to replicate the captured environment accurately. With an emphasis on static physical features, roadways were dissected into relevant road features (RRFs) and full environment (FE) to study the impact of roadside features on the scene complexity and demonstrate the gravity of wildlife-vehicle collisions (WVCs) on AVs. The results indicate that roadside features substantially increase environmental complexity by up to 400%. Increasing a single lane to the road was observed to increase the processing requirements by 12.3-16.5%. Crest vertical curves decrease data rates due to occlusion challenges, with a reported average of 4.2% data loss, while sag curves can increase the complexity by 7%. In horizontal curves, roadside occlusion contributed to severe loss in road information, leading to a decrease in data rate requirements by as much as 19%. As for weather conditions, heavy rain increased the AV's processing demands by a staggering 240% when compared to normal weather conditions. AV developers and government agencies can exploit the findings of this study to better tailor AV designs and meet the necessary infrastructure requirements.

Assessing the Potential for Legal Liability to Create Incentives for Agencies to Reduce Wildlife–Vehicle Collisions in Canada and the United States

Despite proven effectiveness in reducing motorist crashes involving wildlife, systematic inclusion of wildlife-mitigation measures during highway planning and projects continues to vary by jurisdiction in the United States and Canada. Some agencies invest significant resources to reduce wildlife–vehicle collisions (WVCs), while others invest few or no resources. Various factors have been identified as relevant in explaining why some jurisdictions are more proactive than others in addressing WVCs. This analysis builds on existing literature by comparing select legal decisions in the United States and Canada regarding public authority liability for failures to remediate known WVC hot spots, with the goal of assessing the potential role of liability in creating incentives for government decision makers to systematically address motorist crashes involving animals.

Wildlife-vehicle collisions: The disproportionate risk of injury faced by motorcyclists

We applied a generalized linear mixed-effects model to determine the factors leading to injuries from wildlife-vehicle collisions (WVCs). We used the Police database representing WVCs which took place on the Czech road network between 2009 and 2022. The majority of WVCs in Czechia are with roe deer, followed by wild boar, i.e., both relatively small ungulates. Less than 2 % of these encounters ends with an injury to the motor vehicle occupants.We found that the probability of sustaining injury was systematically higher for motorcyclists than for car occupants. The odds of sustaining an injury during WVC were roughly 1600 times higher for motorcyclists than for car occupants. When applying an evading manoeuvre, the odds of sustaining an injury were approximately 68 times higher for car occupants while only 2.3 times higher for motorcyclists compared to a direct hit to an animal. The lack of helmets (for motorcyclists) and missing seat belts (for car occupants) were additional factors which made the outcomes worse for WVCs.While the acceptance of a direct hit (preceded by braking) seems to be a reasonable strategy for car drivers, WVC awareness (including maintaining a lower speed during critical times and places) should be raised among motorcyclists as both manoeuvres are almost comparably dangerous for them.

Traffic and weather influence on small wildlife carcass persistence time on roads

The rapidly expanding road network threatens the persistence of many terrestrial species through mortality associated with wildlife-vehicle collisions. Roadkill monitoring under-estimate actual collision numbers as the time during which roadkill carcasses remain visible on the road is often shorter than the frequency of road monitoring. By placing passerine and amphibian carcasses on stretches of roads we surveyed every 2 h, we fine-tuned existing persistence estimates for these species. The time for half of the carcasses to disappear was less than 30 min for birds and 1–18 h for amphibians depending on the volume of traffic, which is much shorter than previous estimates. We show the implications of our results by performing roadkill surveys on common toads 3 h after the window of reproductive migration, showing roadkill is under-estimated by half, and by estimating from citizen-science data that passerine collisions could be as high as 10 individuals.km-2.year-1 in south-eastern France.

Emerging opportunities for wildlife conservation with sustainable autonomous transportation

Autonomous vehicles (AVs) are expected to play a key role in the future of transportation, and to introduce a disruptive yet potentially beneficial change for wildlife–vehicle interactions. However, this assumption has not been critically examined, and reducing the number of wildlife–vehicle collisions (WVCs) may be beyond current technological capabilities. Here, we introduce a conceptual framework encompassing the intersection between AV technology and wildlife conservation to reduce WVCs. Our proposed framework integrates the development of robust warning systems and animal detection methods for AV systems, and the incorporation of wildlife–vehicle interactions into decision‐making algorithms. With large‐scale AV deployment a looming reality, it is vital to incorporate conservation and sustainability into the societal, ethical, and legal implications of AV technology. We intend our framework to help ecologists and conservationists foster the necessary interdisciplinary collaborations with AV developers and policy makers to reduce WVCs and concomitant biodiversity loss.

Risk Mapping of Wildlife-Vehicle Collisions across the State of Montana, U.S.A.: A Machine Learning Approach for Imbalanced Data along Rural Roads

Abstract Wildlife-vehicle collisions (WVCs) with large animals are estimated to cost the United States over ${\$}$8 billion in property damage, tens of thousands of human injuries, and nearly 200 fatalities each year. Most WVCs occur on rural roads and are not collected evenly among road segments, leading to imbalanced data. There are a disproportionate number of analysis units that have zero WVC cases when investigating large geographic areas for collision risk. Analysis units with zero WVCs can reduce prediction accuracy and weaken the coefficient estimates of statistical learning models. This study demonstrates that the use of the synthetic minority over-sampling technique (SMOTE) to handle imbalanced WVC data in combination with statistical and machine learning models improves the ability to determine seasonal WVC risk across the rural highway network in Montana, USA. An array of regularized variables describing landscape, road, and traffic were used to develop negative binomial and random forest models to infer WVC rates per 100 million vehicle-miles traveled. The RF model is found to work particularly well with SMOTE-augmented data to improve prediction accuracy of seasonal WVC risk. SMOTE-augmented data are found to improve the accuracy to predict crash risk across fine-grained grids while retaining the characteristics of the original dataset. The analyses suggest that SMOTE augmentation mitigates data imbalance that is encountered in seasonally divided WVC data. This research provides the basis for future risk-mapping models and can potentially be used to address the low rates of WVCs and other crash types along rural roads.

Prediction of areas with high risk of roadkill wildlife applying maximum entropy approach and environmental features: East Antioquia, Colombia

Linear infrastructures such as roads are known to cause adverse effects on the surrounding ecosystems. Wildlife–vehicle collisions (WVC) are considered to be one of the main causes of biodiversity loss. Several studies have demonstrated that WVC occurs on Colombian roads. However, studies have focused on a body count, the most affected species, and places with high mortality rates. We aim to propose a methodology for predicting WVC risk in the east of Antioquia, Colombia employing a machine learning approach to identify road segments with a high risk of WVC. Additionally, we present a novel validation technique for the "MaxEnt" approach. During this investigation, 499 reports were collected through road surveys between 2015 and 2016. We identified 160 road segments with high mortality rates with a 2D Hotspots analysis. 15 environmental descriptors were collected for each road segment. Validation of the predictive capabilities of the algorithm was performed using the area under the Receiver Operating Characteristic curve (AUC-ROC). The model achieved a good predictive ability (AUC>0.77). The response curves evidenced that features like distance to forest, land cover, resistance, and land use increase the probability of WVC, specifically, collision risk was higher in zones with high resistance values, crops, and pastures. This methodology has the potential to become an important tool for the prioritization of resources to mitigate WVC.

Social media data from two iconic Neotropical big cats: can this translate to action?

IntroductionThere has been a gradual increase in studies of social media data usage in biodiversity conservation. Social media data is an underused source of information with the potential to maximize the outcomes of established conservation measures. In this study, we assessed how structured social media data can provide insight into species conservation through a species conservation plan, based on predefined actions. MethodsWe established a framework centered on a set of steps that go from defining social media platforms and species of interest to applying general analysis of data based on data dimensions—three W’s framework (What, When, Who) and the public engagement that posts received. The final and most important step in our proposed framework is to assess the overlap between social media data outcomes and measures established in conservation plans. In our study, we used the Brazilian National Action Plan (BNAP) for big cats as our model. We extracted posts and metrics about jaguars (Panthera onca) and pumas (Puma concolor) from two social media platforms, Facebook and Twitter. ResultsWe obtained 159 posts for both jaguars and pumas on Facebook (manually) and 23,869 posts for the jaguar and 14,675 posts for the puma on Twitter (through an application user interface). Data were categorized for content and users (only Facebook data) based on analysis of the content obtained and similarities found between posts. We used descriptive statistics for analyzing the metrics extracted for each data dimension (what, when, who, and engagement). We also used algorithms to predict categories in the Twitter database. Our most important findings were based on the development of a matrix summarizing the overlapping actions and dimensions of the data. Our findings revealed that the most prominent category of information for jaguars on Facebook was the sighting of wildlife outside protected areas, while for pumas, it was the trespassing of property by wildlife. From the Twitter dataset, we observed that the most prominent category of information for jaguars was: the sighting of wildlife outside protected areas, while for pumas, it was wildlife depredation by direct or indirect means. We found temporal trends that highlight the importance of categories in understanding information peaks on Facebook and Twitter. DiscussionWhen we analyze online engagement, we see a predominance of positive reactions on Facebook, and on Twitter, we see a balanced reaction between positive and negative. We identified 10 of 41 actions in the BNAP that might benefit from social media data. Most of the actions that could benefit from our dataset were linked to human–wildlife conflicts and threats, such as wildlife–vehicle collisions. Communication and educational actions could benefit from all dimensions of the data. Our results highlight the variety of information on social media to inform conservation programs and their application to conservation actions. We believe that studies on the success of applying data to conservation measures are the next step in this process and could benefit from input from decision-makers.

Hotspots and Factors Influencing Vertebrate Roadkill on the Ring Changbai Mountain Scenic Road, China

The spatial aggregation patterns of wildlife-vehicle collisions are used to inform where mitigation measures are most needed. Based on 10 years of observations of vertebrate roadkill on the Ring Changbai Mountain Scenic Road, the spatial distribution characteristics of roadkill incidence were analyzed in this study. Using the field survey method to investigate roadkill incidents and their influencing factors, we applied generalized linear mixed modeling (GLMM) for model selection and constructed roadkill models for different taxa groups. The spatial distribution patterns of roadkill hotspots vary among different taxa and exhibit a unimodal or multimodal distribution. The road section along a river and with a minimal distance between the road and the water has a high incidence of roadkill. The density of roadkill for various taxa decreases as the distances from rivers, bridges, and ponds increases. However, there appears to be no correlation between the density of bird roadkill and any factors. Finally, wildlife crossing facilities and guidance measures aimed at reducing the incidence of roadkill and enhancing the selection of routes and wildlife crossing structures in the area are formulated.

Distributional patterns and habitat association of sympatric carnivores in Margalla Hills National Park, Pakistan, and a comparison of conventional versus molecular identification in carnivore scatology

Abstract Knowledge of a species’ distribution is important for developing effective conservation programs. In Pakistan, little is known about the status or distribution of most carnivores coexisting in the same landscape. To address this knowledge gap, we studied distribution patterns of coexisting carnivores in Margalla Hills National Park, using both conventional as well as DNA-identified scats, and other signs. Although scat surveys remain a popular approach to study carnivores, scat identification based on morphology alone is error-prone. As part of our study, we therefore evaluated accuracy of morphological identification of scats using genetic techniques. Field surveys were conducted from September 2015 to December 2018. Using 593 direct (sighting, camera trapping, road kills) and indirect (scats, footprints) field signs, we detected total 11 carnivore species in the park, including 10 that were represented in the subset of 248 (47.7%) scats identified from DNA. The molecular analyses confirmed that the misidentification rate was highest for red fox (Vulpes vulpes) (8.47%), followed by Asiatic jackal (Canis aureus) (7.66%), but least for small Indian civet (Viverricula indica) (3.63%). For investigating habitat association of carnivores, and to test for the associations between species presence and Normalized Difference Vegetation Index (NDVI), we compared the mean and standard deviation of NDVI of each species’ presence locations with the mean and standard deviation of NDVI along the 23 sampling transects. Based on DNA-verified and all scats, carnivore species showed a range of mean NDVI, suggesting, preliminarily, some species may utilize a greater diversity of habitat types than others.

Performance of wildlife fence systems under animal impact load

Wildlife fence protection systems specifically designed for animal protection are critical in mitigating wildlife-vehicle collisions and preserving animal habitats. These systems are quite similar to rockfall barrier systems, which are well studied and developed, however, few studies are specifically oriented toward investigating structural performance of these systems. Despite the similarities, results of studies related to rockfall barriers may not be directly applicable to design and performance evaluation of wildlife barrier systems due to differences in a range of size, shape, and velocity of impactors and structural details. This study aims to bridge this gap by analyzing performance and capacity of wildlife structures in more detail through parametric study with consideration of a wide range of properties and structural details appropriate for these systems, such as mesh type (square and inclined), mesh size, and wire diameters. The behavior of these systems under impact loads is simulated using nonlinear finite element models with consideration of ductile damage and element removal techniques to accurately capture wire rupture and changes in the impactor-fence system interface. The results demonstrate that the response of the system is highly influenced by factors such as mesh type, opening size, wire diameter, and impactor size. Notably, the system utilizing an inclined mesh type exhibits a 35 % higher failure velocity, 50 % greater energy dissipation, and 21 % less maximum displacement compared to the square mesh system. The detailed results and discussions in this study provide valuable insights into the influence of individual parameters on the behavior of wildlife fence systems, aiding in the design of more efficient wildlife barrier systems with enhanced performance.

Carnivores in Peril in the ‘Valley of Extinction’: A Review on Diversity, Distribution Status and Conservation in Southern Assam, India

Southern Assam (in Northeast India), also known as Barak valley, is a part of the Indo-Burma biodiversity hotspot. With some magnificent species already extinct from the region, and some on the verge of extinction, the valley is rightly called the ‘Valley of Extinction’. The region has two wildlife sanctuaries (WLS): Barail WLS and Barak-Bhuban WLS; one proposed WLS (Patharia), and a network of reserve forests (RF) which serve as strongholds of several wild species. However, due to rapid deforestation and decrease in their suitable habitats, encroachment into the wildlife sanctuary and RFs, depletion of prey and food trees, expansion of agriculture, changes in land use and land cover, road kills, etc., most of the wildlife are severely threatened, while others have now adapted to human dominated habitats thereby becoming pests, and vulnerable to prosecution. Moreover, research on the wildlife of the region, excluding a few species, is extremely limited, more so in the case of carnivores owing to their elusive and nocturnal behaviour. The present article focuses on the diversity, status, distribution, threats and conservation of the carnivores of southern Assam. We hereby report a total of 33 species of carnivores from the region, of which 2 are extinct, 2 are probably extinct with only stray individuals. Of the species reported hereby, three are ‘Endangered’, ten are ‘Vulnerable’, and another three are ‘Near threatened’, as enlisted by IUCN. In the beginning of the last century, most of these carnivore species were abundant in southern Assam, which is however decreasing alarmingly. Thus, strict enforcement of Wildlife (Protection) Act, 1972 and Indian Forest Act, 1927; along with awareness campaigns, research on ecology, and population censuses of these species should be carried out to save the remaining populations in the region.

An Analytical Framework of the Factors Affecting Wildlife–Vehicle Collisions and Barriers to Movement

Road mortalities caused by wildlife–vehicle collisions (WVCs) are the most obvious negative effect of roads on wildlife. Identifying the influencing factors and summarizing the spatial-temporal patterns of WVCs have been important research trends in recent decades. However, most studies have only considered a portion of the factors, and there remains a lack of a relatively complete framework, including the numerous factors of WVCs, as well as the underlying transmission mechanisms between factors. In this study, an analytical framework incorporating a wide range of previously discussed factors is constructed. The framework not only displays the possible direction of the influence of each factor on WVCs, but also summarizes some important potential explanations under some circumstances and reveals the main interactions between certain types of factors. From one perspective, the factors affecting WVCs can be divided into four categories: species characteristics, road and traffic characteristics, landscape and environmental characteristics, and driver-related factors and specific human activities. From another perspective, the factors affecting WVCs can be mainly categorized as those related to entering roads and those related to leaving roads safely. The study begins with a discussion of three important sub-frameworks: factors promoting road crossing, factors related to barriers to movement, and factors related to safe crossing. Finally, a suggestion is provided to promote the research on WVCs globally.

Wildlife mortality on roads crossing a protected area: The case of Dadia-Lefkimi-Soufli National Park in north-eastern Greece

Wildlife-vehicle collisions have been identified as an important ecological problem in most countries around the world, during the last decades. In the present work we examined the impacts on wildlife arising from two road categories crossing and adjoining the Dadia-Lefkimi-Soufli National Park (DLS NP), in north-eastern Greece. The study area is consisted of one of the most diverse fauna in Europe, with 202 bird species including 36 out of 38 European diurnal raptors, 60–65 mammal species, 29 reptiles and 13 amphibians. Vehicle collisions was the highest mortality factor in the study (83.9%) among five recorded categories, poison, electrocution, drowning, vehicle collisions and unidentified causes. 204 vehicle casualties were recorded out of 243 in total. Mammals (90.7%) were the vertebrates that mainly suffered from road mortality, including 11 species. The stone marten Martes foina contributed with the highest numbers, followed by the red fox Vulpes vulpes, the hedgehog Erinaceus roumanicus and the wild cat Felis silvestris. Although the contribution of birds on the road casualties was low, nine species were identified, including five diurnal and four nocturnal species, with raptors predominating. Most road casualties were observed in the national road network (71.6%), mainly during holidays (58.3%). Traffic volume, road category, season and blind bend were all dominant factors with different effects determining high wildlife-vehicle collision values. Mitigation measures that can reduce the harmful effects of roads on wildlife in the DLS NP are also discussed.

HUMAN-WILDLIFE CONFLICT AND ROAD COLLISIONS WITH UNGULATES. A RISK ANALYSIS AND DESIGN SOLUTIONS IN TRENTINO, ITALY

Abstract. This study investigates wildlife vehicle collisions with wild ungulates in the Italian Autonomous Province of Trento (PAT) located in the Eastern Italian Alps with a consistent anthropic population and pervasive summer and winter tourism. Both the populations of wild ungulates and vehicular traffic are increasing as well as road collisions leading to animals killed, vehicles damaged and human injuries and fatalities. The purpose of this work was to use FOSS4G to identify the road sections with a high number of collisions and then propose and design practical engineering solutions tailored to each of these hotspots. QGIS 3.16.6, GRASS 7.8.5 and GRASS 8.2 were used to standardize the data set, process georeferenced road collision with ungulates registered by local authorities, perform the hotspot analysis and the final maps. Field surveys were carried out to investigate the local morphology at each hotspot and once the more appropriate practical solutions were chosen, a specific detailed project was proposed including its costs. A cost benefit analysis comparing the cost of the infrastructures with the cost of roadkills shows that the infrastructures are effective in reducing the costs in the medium-long term. The construction of the five proposed infrastructures would reduce deer investments by 6% (about 250 collisions avoided in five years). Such solutions should be more numerous and widely distributed in order to have a greater impact. This FOSS4G procedure can be replicated elsewhere to plan the position of crossing structures, and for application to EU funds, thus mitigating Human-wildlife conflicts (HWC).

On the road again: A study valuing wildlife crossings for wetland mitigation on State Road 40 in Volusia County, Florida

AbstractOver the past few decades, wildlife crossings have effectively reduced wildlife–vehicle collisions and improved habitat connectivity and genetic exchange for animals. However, wildlife crossings are expensive to construct, and the cost may discourage transportation agencies from implementing new wildlife crossings on public roadways. Additionally, transportation agencies and planners rarely consider the permitting and mitigation costs that wildlife crossings can require. A road profile must be raised to accommodate a wildlife underpass on many Florida roadways, particularly in low‐lying areas. Raising the road can result in new wetland impacts that would not occur but for the wildlife crossing. Consequently, new or more compensatory wetland mitigation may be required to satisfy § 404 of the Clean Water Act—a cost to the transportation agency and taxpayers. We used Florida's Uniform Mitigation Assessment Method (UMAM) to calculate the value of functions in habitats before and after planned wildlife crossings on State Road 40 in Volusia County, Florida. The results of our analysis show that wildlife crossings will enhance the functions in remnant habitats adjacent to the wildlife crossings, generating wetland mitigation credits that reduced the amount of compensatory wetland mitigation required by the U.S. Army Corps of Engineers. Our case study provides a novel application of the UMAM in valuing wildlife crossings. We believe the UMAM can be used in cost–benefit models and transportation planning to determine the functional value of new wildlife crossings, thereby producing a monetary value that can incentivize new wildlife crossing projects.

Accounting for cloud cover and circannual variation puts the effect of lunar phase on deer–vehicle collisions into perspective

Abstract Although several studies have focused on the influence of moonlight on deer–vehicle collisions, findings have been inconsistent. This may be due to neglect of the effects of cloud cover, a major impediment to moon illumination and circannual variation in both deer and human activity. We assessed how median cloud cover interacted with the illuminated fraction of the moon in affecting daily roe deer (Capreolus capreolus) roadkill in Slovenia (Central Europe). Data included nationwide roadkill (n = 49,259), collected between 2010 and 2019 by hunters, as required by law. Roadkill peaked under medium to high cloud cover and decreased during nights with low or extremely high cloudiness. This pattern was more pronounced on nights with a full moon. However, the effects of moon illumination and cloud cover had a lower predictive potential than circannual variation, as collisions clearly peaked in April/May, July and August/September. Our results suggest that moonlight could influence roe deer movements through compensatory foraging. However, on nights with a full moon, collisions could also be affected by weather. On bright nights, roe deer might be less active due to increased human presence and sustained vehicular traffic. Then, with medium to high cloud cover and also rainfall, human presence in the environment may be low enough to increase deer movements, but vehicular traffic can still be intermediate, maximizing the risk of collisions. Finally, with overcast skies, widespread rainfall can reduce both traffic volume and human outdoor activity, decreasing the risk of collisions. Moon illumination may indeed affect wildlife–vehicle collisions and roadkill, but its effects should be quantified as a function of cloud cover. Moreover, to make studies truly comparable, research about wildlife–vehicle collisions should also account for time of the year. Policy implications. Because collisions with roe deer peak at particular periods of the year, signs should be installed seasonally. By doing so, they would warn drivers about the risk, improve drivers' awareness and increase their safety. Moreover, as collisions also increase on nights with a full moon and overcast skies, interactive warning signs that are activated by ground illumination should also be useful.

Road Infrastructure and Primate Conservation: Introducing the Global Primate Roadkill Database.

As road infrastructure networks rapidly expand globally, especially in the tropics, previously continuous habitats are being fragmented, resulting in more frequent wildlife-vehicle collisions (WVC). Primates are widespread throughout many sub-/tropical countries, and as their habitats are fragmented, they are increasingly at risk of WVC. We created the Global Primate Roadkill Database (GPRD), the largest available standardized database of primate roadkill incidents. We obtained data from published papers, un-published and citizen science databases, anecdotal reports, news reports, and social media posts. Here, we describe the collection methods for the GPRD and present the most up-to-date version of the database in full. For each primate roadkill incident, we recorded the species killed, the exact location, and the year and month the roadkill was observed. At the time of publication, the GPRD includes 2862 individual primate roadkill records from 41 countries. As primates range in more than twice as many countries, the absence of data from these countries is not necessarily indicative of a lack of primate vehicular collisions. Given the value of these data for addressing both local and global research questions, we encourage conservationists and citizen scientists to contribute to the GPRD so that, together, we can better understand the impact road infrastructure has on primates and evaluate measures which may help mitigate risk-prone areas or species.