Conservation Evidence strives to be as useful to conservationists as possible. Please take our survey to help the team improve our resource.

Providing evidence to improve practice

Action: Install overpasses over roads/railways Terrestrial Mammal Conservation

Key messages

Read our guidance on Key messages before continuing



  • Survival (4 studies): Four studies (including three before-and-after studies), in Canada, Sweden and Australia, found that overpasses (in combination with roadside fencing) reduced collisions between vehicles and mammals. In two of these studies, data from overpasses and underpasses were combined for analysis.


  • Use (21 studies): Nineteen studies, in North America, Europe and Australia, found that overpasses were used by mammals. A wide range of mammals was reported using overpasses, including rodents and shrews, rabbits and hares, carnivores, ungulates, bears, marsupials and short-beaked echidna. A review of crossing structures in Australia, Europe and North America found that overpasses were used by a range of mammals, particularly larger mammal species. A global review of crossing structures (including overpasses) found that all studies reported that the majority of crossings were used by wildlife.

Supporting evidence from individual studies


A replicated study in 1971–1973 of 21 highway overpasses constructed for wildlife use in Québec and Ontario, Canada (Doucet et al. 1974) found that they were extensively used by woodchucks Marmota monax. Woodchucks or their burrows were recorded on 18 of 21 overpasses surveyed. Across four surveys on overpasses, minimum total woodchuck numbers were 16–22. On average, underpasses had 45 woodchucks/100 acres, a high figure compared to those reported by other authors in open flat ground. Twenty-one highway overpasses were built up with rubble and sand and covered with topsoil. Four overpasses had an average area of 72,000 square feet. Overpasses were surveyed once in 1971, twice in 1972 and once in 1973. Surveys were conducted in May, when grass (mainly Agropyron repens) was short. Animals and burrows on overpasses were counted from a vehicle (first two surveys) and on foot (last two surveys).


A study in 1991–1992 along a high-speed railway within agricultural land in Castilla La Mancha, Spain (Rodriguez et al. 1996) found that two flyovers not designed for wildlife were used to cross the railway by small mammals, but not by deer or wild boar Sus scrofa. Small mammals were recorded, with data combined between two overpasses and 15 underpasses, 582 times (37/100 passage-days) and brown hare Lepus granatensis and European rabbit Oryctolagus cuniculus, 89 times (5/100 passage-days). Tracks of four carnivore species, red fox Vulpes vulpes, wild cat Felis silvestris, common genet Genetta genetta and Iberian lynx Lynx pardinus, were recorded. No deer or wild boar Sus scrofa were recorded using overpasses or underpasses. Two flyovers (small roads) crossing a 25-km section of a high-speed railway were monitored. Sand, 3 cm thick and 1 m wide, was put at one entrance to each. Animal tracks were monitored for 15–22 days/month between September 1991 and July 1992.


A replicated study in 1996 of roads in Germany, Switzerland, France and the Netherlands (Keller 1999) found that mammals used flyovers as bridges/overpasses across roads, and frequency of their use tended to increase with overpass width. For all mammal species, frequency of use of the seven narrow overpasses (<15 m wide) was very low. Roe deer Capreolus capreolus used the nine medium-sized (15–50 m wide) and five wide overpasses (>50 m wide) significantly more frequently than they used narrow overpasses. Twenty-one wildlife flyovers/overpasses, in Germany (eight), Switzerland (six), France (four) and the Netherlands (three), were monitored using infra-red video equipment. Flyover widths were 3.4–186 m. Video surveys were carried out during a total of 223 nights.


A replicated, before-and-after study in 1981–1999 in temperate mixed woodland and grassland in Alberta, Canada (Clevenger et al. 2001) found that wildlife overpasses, underpasses and roadside barrier fencing reduced road deaths of large mammals. Species recorded as road casualties included coyote Canis latrans, black bear Ursus americanus, wolf Canis lupus, bighorn sheep Ovis Canadensis, moose Alces alces, deer Odocoileus spp. and elk Cervus canadensis. Mammal-vehicle collisions were significantly lower during the two years after fencing (5–28/year) compared to the two years before (18–93/year) for all three road sections, despite an increase in traffic flow. Ungulate casualties declined by 80%. Most road deaths were within 1 km of the end of the fences. Deaths also occurred close to drainage structures. The Trans-Canada highway was expanded to four lanes and had 2.4-m-high wildlife exclusion fence installed in three phased sections, completed in 1984 (10 km), 1987 (16 km) and 1997 (18 km). In addition, 22 wildlife underpasses and two overpasses were constructed. Wildlife-vehicle collisions were monitored from May 1981 to December 1999.


A study in 1989 and 1994–1995 along a motorway between Arnhem and Apeldoorn in the Netherlands (van Wieren & Worm 2001) found that a wildlife overpass was used by deer, wild boar Sus scrofa, rodents and carnivores. The overpass was used most frequently by red deer Cervus elaphus (1989: 0.1–9 crossings/night; 1994–1995: 4–21) and wild boar (1989: 0.5–21; 1994–1995: 0.5–8.5). It was used less often by roe deer Capreolus capreolus (1989: 2.0 crossings/night; 1995–1994: 0.5) and fallow deer Dama dama (data not presented). Twenty-five rodents and shrews, of three species, wood mouse Apodemus sylvaticus, common vole Microtus arvalis and common shrew Sorex araneus, were caught on the overpass. Overpasses were also used by badger Meles meles and red fox Vulpes vulpes. Overall numbers of crossings was greater in 1994–1995 than 1989 (16 vs 12 crossings/night). The overpass was constructed in the late 1980s. It was 50 m wide, 95 m long and planted with trees. Large mammal tracks were recorded on a 5-m-wide sand strip across the overpass, on 93 occasions in 1989 and 114 occasions in May 1994–April 1995. Small mammals were caught during five nights in summer 1995 using 20 live traps at each end and 32 mouse-traps between.


A replicated study in 1999–2000 in Alberta, Canada (McDonald & St Clair 2004) found that deer mice Peromyscus maniculatus, but not red-backed voles Clethrionomys gapperi or meadow voles Microtus pennsylvanicus, crossed wildlife overpasses. Forty percent of deer mice translocated across roads crossed back over when released alongside overpasses, but no voles did. More animals successfully returned through overpasses (and underpasses) with 100% vegetation cover at entrances (55–100% of animals) compared to those with 50% cover (20–76% of animals) or no cover (0–66% of animals). Those animals that crossed did so in 1–4 days. Two sparsely vegetated wildlife overpasses (75–79 m long, 15 m wide) were used. Territorial mice and voles were caught using Longworth live traps (166 caught in total), ear-tagged, coated with fluorescent powder, translocated across the road, released 2 m from overpasses (or underpasses) and followed as they returned. The amount of ground cover 2 m inside and outside entrances was manipulated to 100%, 50% and no cover, using spruce branches. Traps at original capture sites were monitored for four days after translocation. Animals that did not return were returned by hand. Monitoring was undertaken in July–October 1999 and 2000.


A study in 1997–2000 in Alberta, Canada (Clevenger & Waltho 2005) found that large herbivores and carnivores used two wildlife overpasses. A total of 640 visits to overpasses by elk Cervus canadensis, 1,086 by deer Odocoileus spp., 10 by black bear Ursus americanus, nine by grizzly bear Ursus arctos, eight by wolf Canis lupus and 12 by cougar Puma concolor were recorded, with the majority involving animals crossing the structures. Features that positively influenced use of crossings (two overpasses and 11 underpasses) included increased width, height and openness. Black bears and cougars, though, favoured more constricted crossing structures. Increased length and noise negatively influenced use of crossing structures for some species. Two 50-m-wide overpasses were monitored along an 18-km-stretch of the four-lane Trans-Canada Highway. Barrier fencing, 2.4-m-high, ran alongside the highway. Tracks were monitored at each end of each overpass (in 2 × 4 m of sand/clay), every 3–4 days, from November 1997 to August 2000. Infra-red activated cameras were also used. Information about structure, landscape and human activity were recorded for each overpass.


A study in 2002 in along a road in Zamora, Spain (Mata et al. 2005) found that wildlife overpasses were used by mammals. Overpasses were used by red deer Cervus elaphus (detected at wildlife overpasses on average of 2/10 days), small mammals (shrews, mice and voles; detected 1.0/10 days) and rabbits and hares (detected 4.5/10 days). Other overpasses, such as rural tracks, were used by small mammals (detected 6.4/10 days), rabbits and hares (3.3/10) and foxes Vulpes vulpes (1.4/10), but not by red deer. Two wildlife overpasses (16 m wide, 60 m long) and 16 general overpasses (rural tracks, 7–8 m wide, 58–62 m long) were monitored along a 72-km section of the A-52 motorway. The motorway had barrier fencing along its length. Marble dust (1 m wide cross) was used to record animal tracks for 10 days in June–September 2002. Camera traps were installed on some overpasses.


A global review in 2007 of 123 studies investigating the use of wildlife crossings (van der Ree et al. 2007) found that all studies reported that the majority of underpasses and overpasses were used by wildlife. A total of 1,864 structures were reported on, mainly underpasses (83%; including culverts (742 examples), bridges (130), tunnels (340) and unknown types (333)). Overpasses included land bridges (68), overpasses with small roads (112), canopy bridges (8), glider poles (1) and others (35). Structures provided crossings over or under roads (113 studies), railways (5 studies), both (1 study), canals (2 studies) and a pipeline (1 study). Studies were from Europe (55 studies), the USA (30 studies), Canada (nine studies), South America (one study) and Australia (29 studies).


A study in 2004–2007 in eucalypt woodland in Queensland, Australia (Bond & Jones 2008) found that a wildlife bridge was used by mammals. A total of 1,240 herbivore scats were recorded on the bridge. Brown hare Lepus capensis scats were the most common (78%), followed by red-necked wallaby Macropus rufogriseus (15%), eastern grey kangaroo Macropus giganteus (5%), swamp wallaby Wallabia bicolor (1%), possum (1%) and short-beaked echidna Tachyglossus aculeatus (1%). Six mammals were killed on the road before construction and one afterwards. In 2004, a 1.3-km section of highway was upgraded to four lanes and a variety of wildlife crossings constructed, with barrier fencing (2.5 m high) between. Use of a large overpass (15–20 m wide, 70 m long, planted with grass, shrubs and trees) was monitored from six months after completion. Scats were recorded weekly from August 2005–February 2006 and for two weeks in June 2007. Road-kill was monitored twice weekly before construction (April–July 2004) and weekly afterwards, until June 2007.


A replicated study in 2001 in Zamora province, Spain (Mata et al. 2008) found that overpasses were used by mammals. Wildlife overpasses were used by red fox Vulpes vuples (detected on average per overpass on 3.5/10 days), wild boar Sus scofra (2.3/10 days), small mammals (shrews, mice and voles; 0.3/10 days) and rabbits and hares (3.0/10 days). Other overpasses, such as rural tracks, were also used by wild boar (detected on average per crossing on 0.7/10 days), small mammals (1.0/10 days), rabbits and hares (1.8/10 days), red deer Cervus elaphus (0.2/10 days), rats Rattus sp. (1.3/10 days), western hedgehogs Erinaceus europaeus (0.2/10 days), European badger Meles meles (0.2/10 days) and red fox (3.0/10 days). Cat and dog prints were also detected but could not be determined as being from either wild or domestic species. Overall, overpasses (not including wildlife overpasses) were used disproportionately more than were other crossings (which included underpasses and culverts - data presented as indices). Four wildlife overpasses (15–20 m wide, 60–62 m long) and six general overpasses (rural tracks, 7–8 m wide, 58–65 m long) were monitored along the A-52 motorway. The motorway had barrier fencing along its length. Marble dust (1-m-wide cross) was used to record animal tracks daily for 10 days in March–June 2001.


A before-and-after study in 2002–2004 in mixed forest and farmland in southwestern Sweden (Olsson & Widen 2008, same experimental study site as Olsson et al. 2008) found that following installation of two wildlife overpasses and barrier fencing, moose Alces alces used overpasses and collisions with vehicles decreased, but fencing created a barrier to movements. There were fewer moose-vehicle collisions after overpass and fence construction (zero/year) than before (2.7/year). During construction, 1.8 collisions/year were recorded. Moose were recorded crossing the highway 12 times after overpass and fence installation (during 18 months) and 47 times before installation (eight months). All crossings after construction were via the two wildlife overpasses. Home-range locations changed significantly, with ranges intersected by the highway decreasing to five out of 38 monitored ranges (13%) after fencing from 10 out of 38 (26%) before. Two 6-km sections of the European highway 6 were converted to a fenced four-lane highway in 2000–2004. A third section remained unfenced (3 km). The sections contained two wildlife overpasses, one wildlife underpass, three conventional road tunnels and two conventional bridges that could be crossed. Twenty-four moose were radio-collared. Locations were recorded every two hours before construction (February–September 2002), during construction (October 2002–May 2004) and after construction (June 2004–December 2005).


A before-and-after study in 2000–2005 in forest and farmland in southwestern Sweden (Olsson et al. 2008, same experimental study site as Olsson & Widen 2008) found that a wildlife overpass was used by moose Alces alces and roe deer Capreolus capreolus and, along with barrier fencing, it reduced road-kills. Deaths were reduced 70% from the 12-year pre-construction averages of 2.7 moose killed/year and 5.3 roe deer killed/year. From March 2002–June 2005, the overpass was crossed 437 times by roe deer and 95 times by moose (mainly at night). Roe deer, but not moose crossings, increased over the six-year study. Five to seven individual moose/year used the overpass. Overpass use declined with increased traffic flow. In 2000–2004, a 12-km section of the European Highway 6 was converted from two to four lanes and 2.2-m-high exclusion fencing was installed. Two overpasses and one underpass were constructed. One hourglass-shaped overpass (29–17 m wide, 80 m long, 2 m high, with grey glass-shields to reduce incursion of highway noise and light) was monitored. Tracks were counted in sand beds twice/week and two infrared remote cameras were set overnight. Twenty-four moose were tracked using GPS collars for 22 months.


A site comparison study in 2006 along a highway in New South Wales, Australia (Hayes & Goldingay 2009) found that two wildlife overpasses were used by mammals and presence of crossing-structures along with roadside fencing reduced road-kills. There were fewer road-kills over seven weeks along the section with crossing-structures (0.02/km) than along a section without crossings (0.07/km). The most frequently recorded road casualties along both sections combined were bandicoots (16 casualties) and kangaroo and wallabies (nine casualties). Kangaroos and wallabies used the two overpasses more than they used two underpasses (104 vs 36 tracks). However, the overpasses were used less than were underpasses by bandicoots (28 vs 87) and rodents (15 vs 82). Use was similar for possums (overpasses: 9; underpasses: 14). There were two wildlife bridges (9–37 m wide, with vegetation) and two concrete box culverts (3 × 3 m, 42–63 m long), with 5 km of exclusion fencing, along a 12-km section of dual-carriageway highway. Tracks were monitored on sand plots across each crossing. Road-kill surveys were conducted along the 12-km section and along a 51-km two-lane section without crossings or fencing. Track and road-kill surveys were conducted up to three times/week over seven weeks in August–September 2006.


A study in 2001–2005 along a motorway through forest and agricultural land in Germany (Klar et al. 2009) found that most overpasses, viaducts and underpasses were used by wildcats Felis silvestris to cross roads. Wildcats used crossing structures on 18 of 21 (85%) of the occasions in which they were recorded <50 m from the motorway. Open-span viaducts were used by the highest proportion of cats (five out of seven for which viaducts fell within their home ranges). Forest road overpass were used by one out of eight cats for which road overpasses fell within their home ranges. Two open-span viaducts (335–660 m wide, 29 m long), two forest road overpasses (6 m wide, 46–61 m long) and three underpasses were monitored in 2002–2005. Twelve wildcats were radio-collared between January 2001 and February 2005. Animals were tracked at night for 3–30 months each, to monitor their road crossings.


A study in 1999–2003 along a road through beech and fir forest in Gorski kotar, Croatia (Kusak et al. 2009) found that medium-large mammals used a wildlife overpass (a green bridge) and two other overpasses not specifically designed for wildlife. Monitoring of the green bridge revealed tracks of hare Lepus europaeus (49 tracks), wild boar Sus scrofa (66), roe deer Capreolus capreolus (166), red deer Cervus elaphus (103), fox Vulpes vulpes (83), badger Meles meles (2), brown bear Ursus arctos (39), grey wolf Canis lupus (4) and Eurasian lynx Lynx lynx (1). A similar range of species was recorded on the two other overpasses that were not designed as green bridges (see paper for data). A new highway was constructed in 1998–2004, with 2.1-m barrier fencing. Along a 9-km section, a 100-m-wide green bridge and two overpasses (742 and 835 m wide) above road tunnels, were monitored. Tracks (in snow, mud or sand) and other animal signs were counted 64 times at the green bridge and eight and 23 times at the two other overpasses, in January 1999–January 2001. One of the overpasses was also monitored using a camera trap.


A review of 30 papers monitoring 329 crossing structures in Australia, Europe and North America (Taylor & Goldingay 2010) found that overpasses were used by a range of mammals, particularly larger mammal species. Small mammals used conventional bridge overpasses (demonstrated by 2/4 relevant studies) and wildlife overpasses (4/7 studies). Arboreal mammals used wildlife overpasses (1/1 study). Medium-sized mammals used conventional bridge overpasses (4/5 studies) and wildlife overpasses (5/7 studies). Large mammals used conventional bridge overpasses (9/11 studies) and wildlife overpasses (23/23 studies). Studies suggested that ungulates used overpasses more when they were close to vegetation cover and a river or stream and less when they were in a cropland area. Narrow overpasses (<6 m wide) were not used by deer. Thirty papers, monitoring 329 crossing structures, were reviewed. Fourteen papers investigated multiple structure types, resulting in a total of 52 studies of different structure types. Overpasses included land bridges, wildlife overpasses with grass, trees or other vegetation, combined wildlife and vehicle overpasses, pole bridges and rope bridges.


A replicated study in 2006–2008 of two overpasses over a highway in a Natural Park in Alberta, Canada (Sawaya et al. 2013) found that American black bears Ursus americanus and grizzly bears Ursus arctos used the overpasses. Over three years, a total of eight passages of American black bears (by one individual at each overpass) and 210 of grizzly bears (by 10 individuals at each overpass) were detected. Bear crossings were monitored at two overpasses (dimensions not stated) in Bow Valley, Banff National Park. Overpasses were built in the 1980s and 1990s, and cost >US$2 million each to construct. Bear tracks were counted in May–October 2006, April–October 2007 and April–October 2008 using track pads comprising 1.5–2 m of sandy loam. Track pads were checked every two days and the species, direction of travel, and number of animals was recorded. Individuals were identified by DNA analysis of hairs caught on barbed wires on overpasses.


A review of two studies in 2006–2008 in Australia (Bond & Jones 2014) found that overpasses installed over roads were used by eastern grey kangaroos Macropus giganteus, red-necked wallabies Macropus rufogriseus and swamp wallabies Wallabia bicolor. All road overpasses used fencing to reduce likelihood of animals crossing roads rather than using overpasses. Overpasses in the review were 70 m long and 15 m wide.


A replicated study in 2009 at two sites along a highway through forest in Alberta, Canada (D'Amico et al. 2015) found that North American deer mice Peromyscus maniculatus, southern red-backed voles Myodes gapperi and meadow voles Microtus pennsylvanicus used overpasses to cross a road. Deer mouse tracks were recorded in 75% of track tubes established on overpasses. Southern red-backed vole tracks were detected in 15% and meadow vole in 5% of track tubes. Over two weeks in September–October 2010, small mammals were surveyed on two 50-m-wide wildlife overpasses above the Trans-Canada Highway. Overpasses consisted of sparse young trees, shrubs and open grassland. Two parallel sample lines, each with five 30 cm long × 10 cm diameter track tubes, with sooted metal sheet as a floor, were placed in the centre of each overpass. Mammals were identified from their footprints.


A replicated study in 2010–2014 of five crossing structures at two sites along a highway in Nevada, USA (Simpson et al. 2016) found that more migratory mule deer Odocoileus hemionus used overpasses than underpasses to cross a road. More mule deer crossed the road across two overpasses (234–4,007 deer crossings/overpass/season) than through three underpasses (44–629 deer crossings/underpass/season). Crossing structures, 1.5–2.0 km apart, were located at important crossings for migratory deer. One site had one overpass and two underpasses. The other had one of each structure. Overpasses, made of concrete arches, were 31–49 m wide and 8–20 m long. Cylindrical underpasses were 8 m wide, 28 m long and 6 m tall. All structures had soil bases. Fencing, 2.4 m high, deterred deer from accessing the highway between crossings and extended 0.8–1.6 km beyond crossings at each site. Crossings were monitored, during six to eight mule deer migratory periods (between autumn 2010 and spring 2014) using camera traps, over 10 weeks in each migration (15 September to 1 December and 1 March to 15 May). Cameras were positioned 12 m apart along crossing structures.


A study in 1996–2014 of 18 overpasses and 19 culverts crossing a major highway in Alberta, Canada (Ford et al. 2017) found that overpasses were used by grizzly bears Ursus arctos, particularly in family groups. Over an 18-year period, grizzly bears used overpasses more often (241 crossings/structure) than they used culverts (122 crossings/structure). Over an eight-year period, bear family groups used overpasses more often (1.4 family groups/year/structure) than they used culverts (0.0–0.3 family groups/year/structure). In 1996–2006, 2-m-wide pads, were covered in sandy-loam soil to survey bear movements at 23 crossing structures. From 2008, remote cameras were installed at all crossing structures. As more crossing structures were built in the area, they were added to the survey, up to a maximum of 18 overpasses and 19 culverts. It is not clear when these structures were built.

Referenced papers

Please cite as:

Littlewood, N.A., Rocha, R., Smith, R.K., Martin, P.A., Lockhart, S.L., Schoonover, R.F., Wilman, E., Bladon, A.J., Sainsbury, K.A., Pimm S. and Sutherland, W.J. (2020) Terrestrial Mammal Conservation: Global Evidence for the Effects of Interventions for terrestrial mammals excluding bats and primates. Synopses of Conservation Evidence Series. University of Cambridge, Cambridge, UK.