Aarhus Universitets segl

No. 626: Trafic-kills of larger animals in Denmark

Andersen, P.N. & Madsen, A.B., 2007. NERI Technical Report No. 626, 58 pp.

 

Summary

This project is a co-operative work between the <st1:place w:st="on"><st1:PlaceName w:st="on">Danish</st1:PlaceName> <st1:PlaceType w:st="on">Forest</st1:PlaceType></st1:place> and Nature Agency (DFNA), the National Environmental Research Institute (NERI) and the Danish Hunters’ Association (DHA). The project was initiated by DFNA as part of “Traffic-killed larger animals: Action Plan 2002”. The main objectives have been a precise mapping of traffic-killed larger animals in <st1:country-region w:st="on"><st1:place w:st="on">Denmark</st1:place></st1:country-region> and to identify the patterns of animal-vehicle collisions in order to reduce the number of traffic kills.

 

Tracker dog handlers and selected people from the state forest districts were equipped with GPS-receivers and asked to collect the following information: place of collision, time of collision, age and sex for the following species: roe deer, <st1:City w:st="on"><st1:place w:st="on">red deer</st1:place></st1:City> , fallow deer, sika deer, badger, fox and other species.

 

In the period 1 January 2003 – 31 October 2006, 13,307 traffic-killed larger animals were registered representing 11,711 roe deer, 361 <st1:City w:st="on"><st1:place w:st="on">red deer</st1:place></st1:City> , 57 sika deer, 499 fallow deer, 321 foxes, 143 badgers and 215 other species (Fig. 2). These numbers also include a small number of individuals which were only injured by the traffic. Roe deer is the most frequently recorded traffic killed species (88%) on Danish roads. This is probably due to its relatively large population size, but also because the primary data sources in this project were tracker dog handlers and state forest districts which are normally sent for when a large wild animal is run over. Animal-vehicle collisions peaked in May and the late autumn (Fig. 4). 66% of traffic killed roe deer were females (Fig. 5). Most roe deer-vehicle collisions occurred between 5 and 9 in the morning and between 17 and 23 in the evening (Fig. 6). In winter collisions peaked earlier, between 16 and 20.

 

Landscape characteristics within 500 m from the collision sites, roe deer density and traffic density was analysed by multivariate analysis (logistic regression) to characterize potential hazardous landscape and road stretches. Roe deer density was estimated from annual regional hunting bags. This analysis comprised 3,928 roe deer-vehicle collision sites on roads with known traffic density. Landscape characteristics of collision sites were compared with landscape characteristics on 3,928 randomly selected sites on the road in the vicinity of the collision sites. The result of the total analysis showed that the distribution of traffic killed roe deer can be related to a series of landscape parameters. Their ranked importance was: forest area – roe deer density – traffic density – heather area – length of waterways – wetland and lake area – hedgerow length (Fig. 8). Thus, most of the spatial distribution of roe deer-vehicle collision sites was explained by forest area near the collision site and the length of hedgerows explained the least of the spatial distribution. Only traffic density was negatively correlated with number of roe deer–vehicle collisions.

 

Detailed spatial data was only collected on 10,396 animal-vehicle collision sites (Fig. 9). The total length of public roads in <st1:country-region w:st="on"><st1:place w:st="on">Denmark</st1:place></st1:country-region> in 2005 is 72,257 km, i.e. 0.14 traffic killed large mammal pr. km road. High class roads managed by the national road authorities constitute 6% of the total road network. A relatively high number animal-vehicle collision (16%) was recorded on these roads, i.e. 0.36 traffic killed large mammal pr. km high class road. Only 78 of the 10,396 collisions in <st1:country-region w:st="on"><st1:place w:st="on">Denmark</st1:place></st1:country-region> occurred on road stretches with wildlife fences established before 2003.

 

The regional counties have designated some nature areas as ecological corridors. In the <st1:PlaceType w:st="on">County</st1:PlaceType> of <st1:PlaceName w:st="on">Northern Jutland</st1:PlaceName> 24% of the traffic killed individuals are traffic killed on roads crossing ecological corridors and 29% are traffic killed on roads outside ecological corridors in the <st1:place w:st="on"><st1:PlaceType w:st="on">County</st1:PlaceType> of <st1:PlaceName w:st="on">Northern Jutland</st1:PlaceName></st1:place> (Fig. 14).

 

The density of reported animal-vehicle collision sites were unevenly distributed throughout the country. In the northern and southern Zealand and the northern Jutland and a large area of mid-eastern <st1:place w:st="on">Jutland</st1:place> there are many collisions. In western and southern <st1:place w:st="on">Jutland</st1:place> there are relatively few registrations. These differences probably reflect sampling efforts in the different areas rather than relative population densities of large mammals.

 

The local and regional administrative system in <st1:country-region w:st="on"><st1:place w:st="on">Denmark</st1:place></st1:country-region> was modified by January 2007. The ten new local municipalities with most animal-vehicle collisions are: Frederikshavn, Gribskov, Hillerød, Hjørring, Næstved, Silkeborg, Slagelse, Vejle and Viborg.

 

This co-operation between NGOs, national and regional road and environmental authorities aiming at reducing number of traffic kills of large mammals in <st1:country-region w:st="on"><st1:place w:st="on">Denmark</st1:place></st1:country-region> has successfully located and characterized hazardous road stretches for large mammals.

 

It is concluded that animal-vehicle collisions are concentrated on stretches of roads transverting certain characteristic landscapes rather than being concentrated on distinct smaller “black spots” as originally anticipated in the Action Plan. Although all hazardous road stretches has not been detected by the present study, it is recommended to start constructing mitigation measures for large mammals at the most hazardous road stretches.

 

It is recommended that the collection of detailed data on traffic killed large mammals should be continued. The distribution of traffic killed large mammals may have a different pattern than traffic killings of all mammal species and the project should be extended to include data on other animals to show a more precise picture of number and distribution of traffic killed animals in Denmark.

 

It is suggested that environmental and road authorities in the new local municipalities are encouraged to participate in the project to compile data to support the local authorities’ management of the nature protection regulations, e.g. the protection of Annex IV species in the Habitats Directive.

 

Full report in PDF-format (8,000 kB).