Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
10505574 | Journal of Environmental Management | 2011 | 9 Pages |
Abstract
In addition to posing a serious risk to motorist safety, vehicle collisions with wildlife are a significant threat for many species. Previous spatial modeling has concluded that wildlife-vehicle collisions (WVCs) exhibit clustering on roads, which is attributed to specific landscape and road-related factors. We reviewed twenty-four published manuscripts that used generalized linear models to statistically determine the influence that numerous explanatory predictors have on the location of WVCs. Our motivation was to summarize empirical WVC findings to facilitate application of this knowledge to planning, and design of mitigation strategies on roads. In addition, commonalities between studies were discussed and recommendations for future model design were made. We summarized the type and measurement of each significant predictor and whether they potentially increased or decreased the occurrence of collisions with ungulates, carnivores, small-medium vertebrates, birds, and amphibians and reptiles. WVCs commonly occurred when roads bisect favorable cover, foraging, or breeding habitat for specific species or groups of species. WVCs were generally highest on road sections with high traffic volumes, or low motorist visibility, and when roads cut through drainage movement corridors, or level terrain. Ungulates, birds, small-medium vertebrates, and carnivore collision locations were associated with road-side vegetation and other features such as salt pools. In several cases, results were spurious due to confounding and interacting predictors within the same model. For example, WVCs were less likely to occur when a road bisected steep slopes; however, steep slopes may be located along specific road-types and habitat that also influence the occurrence of WVCs. In conclusion, this review showed that much of the current literature has gleaned the obvious, broad-scale relationships between WVCs and predictors from available data sets, and localized studies can provide unique and novel results. Future research requires specific modeling for each target species on a road-by-road basis, and measuring the predictive power of model results within similar landscapes. In addition, research that builds on the current literature by investigating rare anomalies and interacting variables will assist in providing sound comprehensive guidelines for wildlife mitigation planning on roads.
Keywords
Related Topics
Physical Sciences and Engineering
Energy
Renewable Energy, Sustainability and the Environment
Authors
Kari E. Gunson, Giorgos Mountrakis, Lindi J. Quackenbush,