Factors affecting usage of crossing structures by wildlife – A systematic review and meta-analysis

https://doi.org/10.1016/j.scitotenv.2021.146061Get rights and content

Highlights

  • We reviewed 270 empirical papers on the effectiveness of wildlife crossing structures.

  • Only 28% of global studies reported the proportion of successful crossings to overall approaches.

  • Viaducts are the most effective type of crossing structure for large mammals.

  • Natural materials and round shapes are preferable in the design of crossing structures.

  • Structures designed for dual use are less effective than those designed exclusively for wildlife.

Abstract

Wildlife crossing structures (WCS) are widely used to allow for safe animal movement across roads, promoting both human safety and wildlife conservation. These structures are expensive to build and maintain, and therefore cost-effective design is essential. Although there has been much research to date on the factors affecting the usage of WCS by wildlife, no attempt has been made to synthesize these results and assess the current state of knowledge. We conducted a systematic review of the scientific and professional literature to assess the state of empirical evidence on WCS and a meta-analysis to explore the factors that influence their effectiveness. We identified a total of 270 studies that reported empirical results spanning four decades of research. Most studies (161) measured the number of crossing events without monitoring approaches to the structure, thus limiting the ability to assess WCS effectiveness. Only 77 studies measured the proportion of successful crossings to approaches, which was the type of data used for meta-analysis.

Our meta-analysis results show that viaducts are the most effective type of WCS for large mammals. For example, the odds of ungulates crossing through a viaduct are 2.9 times that of an overpass, and 3.6 times that of an underpass. WCS built specifically for wildlife are used significantly more than those built for dual use by humans and wildlife. For large carnivores, the odds of using a dedicated WCS are 15.9 times that of a structure used concurrently by humans. We additionally found that natural materials and round shapes are preferable in the design of effective WCS. Altogether, these results highlight the importance of adopting large-scale monitoring of wildlife crossing structures. More broadly, we conclude that further research focusing on under-studied species and structure characteristics is needed to facilitate cost-effective mitigation efforts that reduce wildlife-vehicle collisions and promote wildlife conservation.

Introduction

The global transportation infrastructure network has been expanding rapidly in the past century, and is expected to expand much further in the coming decades (Laurance et al., 2014). Transportation infrastructures such as roads and railroads are recognized as a major threat to global biodiversity (Forman and Alexander, 1998; Torres et al., 2016). Roads act as a barrier to wildlife movement within and between habitats and increase mortality due to vehicle collisions (Fahrig and Rytwinski, 2009). Some wildlife species tend to avoid the unnatural surface of the road and the disturbances associated with roads such as moving vehicles, noise, and light (D'Amico et al., 2015a). Roads thus reduce connectivity, fragmenting animal populations into small subpopulations that are vulnerable to local extinction, threatening their long-term viability, and ultimately decreasing biodiversity (Rytwinski and Fahrig, 2015; Bennett, 2017). Much effort has been put forth to mitigate these effects (van der Ree et al., 2007). It is becoming increasingly common to install wildlife crossing structures (WCS) allowing for safe animal movement, improving human safety, and reducing property damage (Clevenger and Waltho, 2005; Forman et al., 2003; van der Grift et al., 2013). WCS are designed to facilitate movement of animals, connect populations, and reduce wildlife mortality (Corlatti et al., 2009). There are multiple types and designs of structures, from small culverts to open-span viaducts and large overpasses. Globally, hundreds of wildlife overpasses and underpasses have been constructed in the past 50 years, most commonly in Europe and North America, and the rate of construction has been steadily rising over time (Seidler et al., 2018). Although there has been much research to date on the factors affecting the usage of crossing structures in specific case studies, no attempt has been made to synthesize these results and assess the quality of research and the current state of knowledge. Here, we attempt to address these knowledge gaps using a systematic review and meta-analysis.

Wildlife crossing structures are expensive to build and maintain, costing up to millions of dollars for a single overpass, and global costs of these mitigation measures exceed hundreds of millions of dollars (Ascensão and Mira, 2007; Glista et al., 2009). Therefore, these structures should be cost-effective and designed to allow for maximal use by target species. Studies on the usage of WCS have explored the variation in species preferences for crossing structure characteristics. However, although preferences vary across species, some studies revealed that differences in use patterns are more pronounced between functional species groups, delineated by body size and ecological function (Clevenger and Waltho, 2000; Ascensão and Mira, 2007). Several studies have focused on the impact of structure dimensions (i.e., length, width, height, and openness), particularly for large mammals, demonstrating negative effects of structure length on use by ungulates (Ng et al., 2004; Clevenger and Waltho, 2005; Wang et al., 2018). For large and small carnivores, increasing WCS width was shown to have a positive effect (Craveiro et al., 2019; Serronha et al., 2013; Grilo et al., 2008; Mata et al., 2003), but these results were not consistent (Seiler and Olsson, 2009). Other structural attributes including shape, substrate and construction material have not been widely studied, although some studies have highlighted that these attributes can also influence use patterns (Smith, 2003). For example, it was shown that salamanders prefer a sandy substrate over bare concrete (Patrick et al., 2010). Studies of herpetofauna are scarce and show mostly insignificant results, although structure length has been shown to have negative effects for reptiles in some examples (Ascensão and Mira, 2007; Woltz et al., 2008). The large variation in responses to WCS characteristics poses considerable challenges to WCS planning and requires a better understanding of this interspecific variation.

Beyond WCS characteristics, the environment surrounding the structure can also play a role in the level of use by wildlife. Human activity near the structure was studied in several cases and results demonstrate conflicting responses to human presence, regarding both ungulates and large carnivores (Ng et al., 2004; Clevenger, 1998; Wang et al., 2018; Georgii et al., 2011; Grilo et al., 2008; Clevenger and Waltho, 2000). Fencing is considered as one of the key road mitigation measures that is used to prevent animals from accessing roads and direct movements towards WCS (van der Grift et al., 2013). Fencing appears to promote ungulate use of underpasses (Dodd et al., 2007; Huijser et al., 2016). Other attributes of the surroundings were studied, for instance vegetation cover at the entrance of the structure, with mostly conflicting results as in the case of small non-carnivores showing both negative and positive effects (Rodriguez et al., 1996; Grilo et al., 2008; Smith, 2003). Differences between the usage of various structure types has been explored by few studies, mostly for small mammals, for which a preference for overpasses over culverts has been found (Mata et al., 2003). These results regarding the effects of structural and environmental factors on usage reflect the risk-disturbance hypothesis. It has been suggested that risk-avoidance responses related to anthropogenic stimuli cause deviations in behavior relative to that expected without human influence (Frid and Dill, 2002). Thus, use of structures that allow for safe passage is diminished when the structure characteristics and surroundings are perceived as risky by wildlife.

Many studies on WCS have documented the use of crossing structures by recording the number of animals that cross through a structure. Researchers use motion cameras, track beds, radiotracking and human observations to identify successful crossing events (van der Ree and Tonjes, 2015). While this is an important measure for assessing functionality of each structure as a conduit for local wildlife movement, the number of crossing events alone provides little information on a structure's effectiveness in comparison with other possible structure designs (Chambers and Bencini, 2015). Wildlife population densities and the permeability of the surrounding landscape influence the number of approaches to individual structures. If the number of approaches to the structure is not measured, it is unknown whether the number of crossing events constitutes a large or small proportion of the individuals that could have potentially used the structure (Clevenger, 2011). Hence, conclusions from such studies might be biased regarding the effects of structure attributes on the expected usage by wildlife and are difficult to generalize. Studies that did report the number of approaches as well as successful crossings have used the same tracking methods as mentioned above, but at a larger area surrounding the crossing structure (Andis et al., 2017). This method allows for the identification of individuals that explore the area of the structure but avoid using it to cross through.

Planning of cost-effective WCS requires unbiased knowledge of the structural and environmental factors that affect usage by different taxa, so that funds can be effectively directed to build structures which will most likely be used by target species. In this study, we aimed to understand the current state of empirical literature regarding the effects of WCS use by wildlife. We ask how much of the current reported results in the literature are potentially biased due to lack of monitoring of approaches to the crossing structure. To answer this question, we carried out a systematic review of both the scientific and professional literature to map existing knowledge gained in the last four decades of empirical research. We also aimed at uncovering understudied and overarching effects of WCS characteristics on usage by six functional groups: ungulates, large carnivores, small carnivores, small non-carnivores, herpetofauna and macropods. Finally, we asked which WCS characteristics contribute to the effectiveness of crossing structures, and how do these effects vary among studies and WCS types. To this end, we studied the effects of multiple factors on the proportion of successful crossings out of overall approaches to the structure (PSC) by using a meta-analysis of the empirical data found in the literature. Based on the current literature, we expected to find that wider and shorter crossing structures would be more effective for most species groups, and that fencing will have a positive effect on usage. We also expected that viaducts and overpasses would be more effective than underpasses for large species, and that human use of WCS will have a detrimental effect on the levels of usage by wildlife species.

Section snippets

Methods

We carried out a systematic review (Hillebrand and Gurevitch, 2016) to identify all the existing literature in the field of wildlife crossing structures. First, we devised a protocol outlining the methods that will be used to conduct the systematic review. The protocol specifies the search terms, relevant databases, inclusion and exclusion criteria, data collection methods and data synthesis methods (Supporting information, Text S1). The protocol was sent to seven experts in the fields of road

Systematic review

The number of papers found in all databases was 10176, out of which 1127 were left after removing papers with a title that was obviously irrelevant for the research topic and removing duplicates that appeared in multiple searches. Out of these 1127 papers, the full text of 41 was non-accessible and they were discarded. For the remaining 1086 papers the abstracts were read. Based on the abstracts, 270 papers that met the inclusion criteria were identified (see Supporting information, Table S7

Current state of the literature

Transport infrastructures are increasingly recognized as one of the major drivers of biodiversity loss worldwide (Benitez-Lopez et al., 2010; Polak et al., 2019). Mitigation measures, such as WCS are gaining attention by transport agencies, because they can allow species movements across transport infrastructures and prevent wildlife-vehicle collisions (van der Grift et al., 2013). Yet, underpinning research is still needed to identify best practices and ensure funds are allocated in a

Conclusions

Our systematic review results highlight the importance of adopting large-scale monitoring of wildlife crossing structures, so that probabilities of use by approaching individuals may be measured. This metric of crossing structure effectiveness should become the standard for future research in the field. To this end, we recommend the development of standardized monitoring schemes and methods, adjusted for specific species groups, so that structure effectiveness could be measured and compared

CRediT authorship contribution statement

Dror Denneboom: Conceptualization, Methodology, Software, Validation, Formal Analysis, Investigation, Data curation, Writing - Original draft preparation, Writing - Review & Editing, Visualization. Avi Bar-Massada: Conceptualization, Methodology, Writing - Review & Editing. Assaf Shwartz: Conceptualization, Methodology, Writing - Review & Editing, Supervision.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

We would like to thank the experts that have provided suggestions for the improvement of the protocol for the systematic review and meta-analysis: Dr. Christina Mata, Dr. John A. Bissonette, Dr. Marcel Huijser, Prof. Yohay Carmel, Dr. Orr Spiegel, Dr. Tal Polak, and Mr. Dotan Rotem. We would also like to thank the anonymous reviewers who have helped us improve this research paper.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data availability statement

Data is available from Figshare at https://doi.org/10.6084/m9.figshare.12623756.

References (55)

  • H.W. Woltz et al.

    Road crossing structures for amphibians and reptiles: informing design through behavioral analysis

    Biol. Conserv.

    (2008)
  • A.Z. Andis et al.

    Performance of arch-style road crossing structures from relative movement rates of large mammals

    Front. Ecol. Evol.

    (2017)
  • F. Ascensão et al.

    Factors affecting culvert use by vertebrates along two stretches of road in southern Portugal

    Ecol. Res.

    (2007)
  • F. Barraquand et al.

    Animal movements in heterogeneous landscapes: identifying profitable places and homogeneous movement bouts

    Ecology.

    (2009)
  • D. Bates et al.

    Fitting linear mixed-effects models using lme4

    J. Stat. Softw.

    (2015)
  • V.J. Bennett

    Effects of road density and pattern on the conservation of species and biodiversity

    Curr. Landsc. Ecol. Rep.

    (2017)
  • P.D. Chachelle et al.

    Western grey kangaroos (Macropus fuliginosus) include fauna underpasses in their home range

    Wildl. Res.

    (2016)
  • B. Chambers et al.

    Factors affecting the use of fauna underpasses by bandicoots and bobtail lizards

    Anim. Conserv.

    (2015)
  • A.P. Clevenger

    Permeability of the Trans-Canada highway to wildlife in Banff National Park: importance of crossing structures and factors influencing their effectiveness

  • A.P. Clevenger

    15 years of banff research: what we’ve learned and why it’s important to transportation managers beyond the park boundary

  • A.P. Clevenger et al.

    Factors influencing the effectiveness of wildlife underpasses in Banff National Park, Alberta, Canada

    Conserv. Biol.

    (2000)
  • Clevenger, A.P., Ford, A.T., Sawaya, M.A. (2009). Banff wildlife crossings project: integrating science and education...
  • L. Corlatti et al.

    Ability of wildlife overpasses to provide connectivity and prevent genetic isolation

    Conserv. Biol.

    (2009)
  • D'Amico, M., Clevenger, A.P, Roman, J., Revilla, E. (2015a). General versus specific surveys: estimating the...
  • T.P. Debray et al.

    Individual participant data meta-analysis for a binary outcome: one-stage or two-stage?

    PLoS One

    (2013)
  • N.L. Dodd et al.

    Role of fencing in promoting wildlife underpass use and highway permeability

  • L. Fahrig et al.

    Effects of roads on animal abundance: an empirical review and synthesis

    Ecol. Soc.

    (2009)
  • Cited by (33)

    • Evaluation of motorists perceptions toward collision of an endangered large herbivore in Iran

      2023, Global Ecology and Conservation
      Citation Excerpt :

      Wildlife-vehicle collision (hereafter WVC) is an intrinsically ecological and social issue (Crawford and Andrews, 2016). While most of WVC research has focused on the biophysical effects of roads on wildlife (Nguyen et al., 2021; Denneboom et al., 2021, 2021c), research on human dimensions that influence WVC is not extensive (Crawford and Andrews, 2016; van de Water and Matteson, 2018). Evaluating motorist attitudes toward WVC and knowledge of wildlife, especially endangered species, has the potential to increase our understanding of the social factors that affect WVC, as well as the broader impacts of roads on wildlife to develop effective mitigation and education strategies (Crawford and Andrews, 2016).

    • Crossing structure use in a tiger landscape, and implications for multi-species mitigation

      2022, Transportation Research Part D: Transport and Environment
      Citation Excerpt :

      While increasing widths decreased chital, tiger, wild dog and wild pig use of CS, the highest significant threshold estimates for CS width was calculated at 300 m (gaur). Wild animals are known to change their activity patterns to avoid interacting with humans (Gaynor et al., 2018), and the odds of wildlife using CS with human co-use are less than structures dedicated for wildlife use (Denneboom et al., 2021). While human use was not a significant influence on CS use by wildlife, temporal patterns of avoidance of human activity were observed in most species, with the exception of chital, wild pigs and wild dogs that displayed similar activities under different levels of human use.

    View all citing articles on Scopus
    View full text