Loading...
Please wait, while we are loading the content...
Similar Documents
Title Evaluating when and where to implement road mitigation for wildlife with roadkill modelling : three international case studies
| Content Provider | Semantic Scholar |
|---|---|
| Author | Teixeira, Fernanda Zimmermann Gunson, Kari |
| Copyright Year | 2017 |
| Abstract | Roads pose many threats to wildlife including mortality resulting from wildlife-vehicle collisions (WVC). Although considerable research on WVC exists, analysing and predicting risk with statistical modelling has not been fully applied in management or road planning. Models can augment these efforts by identifying or predicting high-risk areas across time and space. However, for models to be effectively incorporated into management practice, they must be conceptually simple, flexible to changing data or environments, and adaptable for a range of ecological problems or species. We present three case studies on different continents where modelling is used to describe and predict WVC risk. In Canada, road mortality is an identified threat for seven species of turtles. Two years of road mortality was documented along 100 km of highway that bisects extensive wetlands in Eastern Ontario. WVC hotspots were identified and the transportation agency is now installing permanent exclusion fencing where the highest peak of turtle mortality occurred. Subsequently, the model was extrapolated to the provincial road network and its prediction capability validated with an independent data set. In Brazil, modelling is used to identify factors related to spatial and temporal distributions of amphibian roadkill in the Atlantic Forest Biosphere Reserve. The results suggest locations and seasons for placement of mitigation. In Australia, WVC are predicted with an adapted conceptual risk framework by modelling the magnitude of threat (vehicle presence and speed) and exposure to threat (wildlife presence) across the State of Victoria for ten species of mammals and birds. Reported locations of WVC are used to train and validate a collision model and measure which predictors influence risk. Managers can then manipulate predictor values, such as speed limit or traffic volume, in a simulated environment, and observe predicted collision risk on road networks. While employing different analytical mechanisms, all case studies introduce novel methods to pro-actively reduce WVC. |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | https://www.ufrgs.br/nerf/images/documentos/Abstract_ICCB_2015.pdf |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |
