Collisions between vehicles and wildlife are a serious problem in many countries around the world. Ungulate-vehicle collisions are the most problematic, because the damage from such collisions, including injuries and fatalities of humans and animals, tends to be serious. The number of sika deeron Hokkaido, the northernmost island of Japan, is steadily increasing, asare deer-vehicle collisions (DVCs) there. To achieve greater traffic safety and harmony between wildlife and humans by developing measures to prevent such collisions, this study investigated the details of DVCs on the seven national highways in the Tokachi area, one of the most DVC-prone area in Hokkaido, using a database that includes the number of DVCs for the 12 years from 1995, each DVC location, the route name and the pickup date. First, the relationship between DVC frequency and deer seasonal behavior was investigated. Next, a Poisson regression model was used to understand the effect of explanatory valuables on the number of DVCs. These explanatory valuables included DVC site topography features such as avg. annual max. snow depth, avg. elevation, vegetation type, land-use classification, distance from the nearest river and average day traffic volume (ADT). A topographic dataset including the spatial distribution of avg. elevation and of avg. annual max. snow depth, and landscape features within 1,000-m-radius zones around each kilopost on each sample route was created by using GIS software. Those analysis results identified the specific conditions of DVC-prone sites. For example, it was found that collisions tended to be more common in the spring migratory season and the autumn breeding season, and on road sections through woodland or those in a border area between woodland and farmland. These results provide useful information for identifying what DVC prevention measures should be implemented and where they should be implemented.
Samenvatting