Skip to main content
SLU publication database (SLUpub)

Conference paper2009Peer reviewedOpen access

Are non-wildlife passages effective passages for wildlife?

Seiler, Andreas; Olsson, Mattias


In order to mitigate barrier effects of highways and exclusion fences on wildlife, many countries have invested in specific wildlife crossing structures placed at selected strategic locations. While such structures may be significant to species conservation or management at local scale, they may not necessarily suffice to maintain landscape connectivity at broad scale. Conventional, non-wildlife road bridges, tunnels and culverts, however, are usually abundant along the major infrastructure corridors and are known to be used by animals at least occasionally. Given the large number and density of such passages, their accumulative effect may well be underestimated. On the other hand, there is uncertainty about how effectiveness of wildlife passages should be judged, because clear objectives and performance targets are undeveloped. We used track inventories to study the relative use of a total of 57 conventional road underpasses in south-central Sweden by common wildlife species such as moose (Alces alces), roe deer (Capreolus capreolus), red fox (Vulpes vulpes), badger (Meles meles) and hares (Lepus spp.). We studied the influence of passage dimensions, design, human disturbance and landscape factors and derived recommendations on limits in size and openness based on selected multiple regressions. Our results support earlier findings in that ungulates are more sensitive to underpass dimensions as medium-sized carnivores and hares. In general, moose and, to some degree also roe deer, used underpasses much less than expected from their occurrence in the surrounding habitat, whereas badgers and foxes, in particular, showed clear preference towards the underpasses. Openness appeared as a strong predictor for the relative use by most species, but also traffic within the underpasses and distance to nearest forest cover were important variables. Landscape attributes, such as habitat composition within 500 m around the passage or the distance to the nearest alternative crossing option, were of less significance to the relative use of underpasses. We estimated that underpasses with a relative openness of 2.3 (and minimum width of 11m), with limited human and vehicular traffic (12 passages per day) and nearby forest cover (distance <15 m) are likely to be used by moose at random, i.e., as much as expected from moose activity on control track beds. Smaller animals, including roe deer, will use such passages more frequently. We propose establishing random passage use (use as expected) as a performance target for non-wildlife crossing structures. Higher targets should be set for adapted wildlife passages. Additional, ecologically scaled performance targets must address the distance between adjacent crossing facilities. We conclude that, at least in Sweden, only a minor proportion of conventional road underpasses built for local access roads provide effective passages to roe deer and smaller species, and only very few to moose. It is worthwhile studying, however, whether other facilities can be created to provide safe passage for wildlife across roads or whether additional protective features can increase the attractiveness of existing structures and thereby provide more costefficient mitigation than the investment in new, adapted wildlife passages.


Sweden; road underpasses; wildlife crossing structures; wildlife passages

Published in


International Conference on Ecology & Transportation

    UKÄ Subject classification

    Behavioral Sciences Biology

    Publication identifier


    Permanent link to this page (URI)