Action: Install tunnels/culverts/underpass under roads
Key messagesRead our guidance on Key messages before continuing
- Twenty-five studies evaluated the effects on mammals of installing tunnels, culverts or underpass under roads. Eight studies were in the USA, four were in Australia, four were in Canada, two were in Spain, one each was in Germany, the Netherlands and South Korea and three were reviews with wide geographic coverage.
COMMUNITY RESPONSE (0 STUDIES)
POPULATION RESPONSE (3 STUDIES)
- Survival (3 studies): A study in South Korea found that road sections with higher underpass density did not have fewer wildlife-vehicle collisions. A review found that most studies recorded no evidence of predation of mammals using crossings under roads. A controlled, before-and-after, site comparison study in Australia found that overwinter survival of mountain pygmy-possums increased after an artificial rocky corridor, which included two underpasses, was installed.
BEHAVIOUR (23 STUDIES)
- Use (23 studies): Seventeen of 20 studies (including seven replicated studies and two reviews), in the USA, Canada, Australia, Spain, the Netherlands, and across multiple continents, found that crossing structures beneath roads were used by mammals whilst two studies found mixed results depending on species and one study found that culverts were rarely used as crossings by mammals. One of the studies found that crossing structures were used by two of four species more than expected compared to their movements through adjacent habitats. A controlled, before-and-after, site comparison study in Australia found that an artificial rocky corridor, which included two underpasses, was used by mountain pygmy-possums. A replicated study in Germany found that use of tunnels by fallow deer was affected by tunnel colour and design. A study in the USA found that a range of mammals used culverts, including those with shelves fastened to the sides.
- Behaviour change (1 study): A controlled, before-and-after, site comparison study in Australia found that after an artificial rocky corridor, which included two underpasses, was installed, dispersal of mountain pygmy-possums increased.
Tunnels, culverts and underpasses may provide safe road crossing opportunities for mammals. A range of different tunnels can be used, including purpose-built wildlife tunnels, culverts that assist with drainage and which can also be used by wildlife, and large passages beneath elevated road section which may sometimes also be used for local vehicle access.
Underpasses are frequently installed in conjunction with wildlife barrier fencing which funnels animals towards the tunnel and prevents them from accessing the road. For this combined intervention, see Install barrier fencing and underpasses along roads. See also Install tunnels/culverts/underpass under railways.
Studies included here are those where barrier fencing is not installed or not explicitly referred to in the study methods or where at least some underpasses were in unfenced areas. Most studies here report solely on the use of these structures, such as the number of crossings made. There is an absence of studies reporting on wider population-level effects of the presence of these structures.
Supporting evidence from individual studies
A controlled, before-and-after, site comparison study in 1982–1986 of rock screes and boulder fields on a mountain in Victoria, Australia (Mansergh & Scotts 1989) found that an artificial rocky corridor, which included two underpasses, was used by mountain pygmy-possums Burramys parvus and female overwinter survival and male dispersal increased. Over 28 days, mountain pygmy-possum were recorded in a monitored underpass 60 times, bush rats Rattus fuscipes 21 times and dusky antechinus Antechinus swainsonii three times. The overwinter survival of female pygmy-possums was 96% of that at an undisturbed site after corridor construction, compared to 21% before. Before construction, sex ratios at the two sites differed, with males not dispersing at the developed site. After construction, both adult and juvenile males dispersed (population before: 25% male; after: 10% male). In 1985, a 60-m-long corridor, connecting a fragmented breeding area, was created. This included two adjacent tunnels (1 m diameter) under a road. The corridor and tunnels were filled with rocks to imitate scree. A remotely activated camera monitored one tunnel over 18 days in February–April and 10 days in October–November 1986. Possums were live-trapped in 1982–1986. Population composition was compared at the developed (ski resort) site and one undisturbed site.
A replicated study in 1994 of tunnels in enclosures in Germany (Woelfel & Krueger 1995) found that use of tunnels by fallow deer Dama dama was affected by tunnel colour and design. Deer used one tunnel significantly more in four of six paired trials. A white-painted tunnel was used more than a grey-painted tunnel (732 vs 425 passages) and also more than a black-painted tunnel (294 vs 153 passages). A black base was used more than one without a base (747 vs 584 passage). An unlit tunnel was used more than an indirectly-lit tunnel (581 vs 242). There was no significant difference in the use of tunnels with and without tree stumps within them. Two tunnels were erected in a 0.7-ha enclosure, each 2 m high, 2 m wide and 8 m long. Twenty deer accessed food through the tunnels. Tunnel use was registered by a photo-electric sensor. Trials were run with six tunnel design combinations: both tunnels unpainted; white vs grey; white vs black; black base (and 80 cm up sides) vs no base; indirect light on ceiling vs unlit; tree stumps in tunnel vs no stumps. Tunnels were painted off-white for base, lighting and tree stump trials.
A replicated study in 1994 of roads and railways in Madrid province, Spain (Yanes et al. 1995) found that all 17 culverts under roads were used by mammals. The highest frequencies of tracks were from wood mice Apodemus sylvaticus (2.5 tracks/day), shrews Sorex spp. (0.5/day) and European rabbits Oryctolagus cuniculus (0.3/day). Rats Rattus sp. (0.1 tracks/day), hedgehogs Erinaceus europaeus (0.01/day), cats (mostly wild cat Felis sylvestris - 0.04/day), red fox Vulpes vulpes (0.03/day), genet Genetta genetta (0.02/day) and weasel Mustela nivalis (0.01/day) were also detected. Small mammal use of culverts decreased with increased road width and culvert length and increased with increased culvert height, width and openness. Use by rabbits and carnivores decreased with increasing highway or railway width. Rabbit use also declined with increased boundary fence height (fences ran across culvert entrances, rather than funnelling animals towards them). Vegetation complexity had little influence. Five culverts were monitored under railways, two under a motorway and 10 under local roads. Structural, vegetation and traffic variables were recorded at each culvert. Use was monitored using marble (rock) dust over culvert floors to record tracks. Sampling was undertaken in 1994, over four days each in spring, summer, autumn and winter. Sampling extended to eight days at four culverts when deer were nearby.
A replicated study in 1993–1994 along four roads in Catalonia, Spain (Rosell et al. 1997) found that underpasses were used by several mammal species. Small mammals used all rectangular culverts and 94% of circular culverts. Hares Lepus spp. and rabbits Oryctolagos cuniculus used 83% and 23% of rectangular and circular culverts respectively whilst carnivores used 88% and 75% respectively. Carnivores recorded were weasel Mustela nivalis, beech marten Martes foina, badger Meles meles, genet Genetta genetta and fox Vulpes vulpes. Wild boar Sus scrofa and roe deer Capreolus capreolus also used underpasses. Use was greater by small mammals for underpasses at the same level as the surroundings and those with natural substrate on the floor. Those with water were used less frequently. Rabbits did not use narrow structures (<1.5 m), whereas wild boar used underpasses >7 m wide. A total of 39 circular (1–3 m diameter) and 17 rectangular drains (4–12 m wide) and other underpasses were surveyed along four 10-km sections of road. Underpasses were monitored for four days/season over a year, in 1993–1994. Animal tracks were monitored using marble power (50 cm wide) across the centre of each structure. Infra-red and photographic cameras were used at entrances.
A study in 1996–1997 along a highway in New South Wales, Australia (Norman et al. 1998) found that mammals used three underpasses. Between three and nine native mammal species used each of the tunnels. Common wombat Vombatus ursinus, swamp wallaby Wallabia bicolor, rats (Rattus fuscipes, Rattus lutreolus) and bandicoots (Perameles nasuta, Isoodon macrourus) were the most frequently recorded. Four non-native species also used underpasses. The greatest number of species was recorded in the largest underpass, but the smallest underpass had the greatest frequency of use. A total of 43 native and 57 introduced mammals were killed on the road during the survey. Three underpasses (diameters: 1.5–10 m) were monitored from August 1996 to June 1997. Infra-red camera traps, track counts (sand 2 m inside entrances), trapping and nocturnal searches were used. Road-kill data were also collected.
A replicated study in 1997–1998 of 53 wildlife passages along waterways under roads at over 20 sites in the Netherlands (Veenbaas & Brandjes 1999) found that all passages were used by mammals. At least 16 mammal species used passages. Waterside banks extending under bridges were used by 14 species and other types of passageways by 10 species. Brown rats Rattus norvegicus, mice and voles were the most frequently recorded mammals (see original publication for details). For all mammals, frequency of use increased with increasing passage diameter and width, but was not affected by substrate. Culverts and bridges were adapted for wildlife, in the 1990s. In 1997, thirty-one passages (0.4–3.5 m wide) were monitored. These included extended banks (unpaved or paved), planks along bridge or culvert walls, planks floating on the water, concrete passageways and plastic gutters covered with sand. In 1998, twenty-two passages were monitored for the effect of width and substrate. These were wooden passageways along bridge or culvert walls (0.2–0.6 m wide). Monitoring involved weekly checks of tracks on sandbeds (for 4–7 weeks) and ink pads (12 weeks in 1997, four weeks in 1998) across passageways.
A study in 2000 along a highway in Vermont, USA (Austin & Garland 2001) found that a concrete underpass was used by four mammal species to cross the road. Infra-red monitors recorded 190 confirmed or unconfirmed instances of animals using the tunnel. Where a species was identified, 58% of occurrences were racoon Procyon lotor, 27% were mink Neovison vison, 11% were weasel Mustela frenata and 4% were skunk Mephitis mephitis. The total number of passages by these species was not stated. The underpass was a concrete block structure, split along the middle by a concrete support. It was 97 m long, 3 m wide and 4 m high. A stream flowed through one tunnel and, at times of high water, through both tunnels, though a sloping floor ensured at least some dry passage. The underpass was monitored discontinuously from June–November 2000, using infrared monitors, cameras and footprint pads.
A replicated study in 1999–2000 along two highways in Alberta, Canada (Clevenger et al. 2001) found that drainage culverts were used by at least nine mammal species. A total of 618 crossings were recorded. Species recorded were coyote Canis latrans (1% of crossings), American marten Martes americana (12%), weasel Mustela ermine and Mustela frenata (28%), snowshoe hare Lepus americanus (3%), red squirrel Tamiasciurus hudsonicus (4%), bushy-tailed wood rat Neotoma cinerea (15%), shrew spp. Sorex spp. (8%), deer mouse Peromyscus maniculatus (28%) and vole spp. Arvicolinae (0.5%). Culvert use was positively correlated with traffic volume (for hare, squirrel and marten), culvert openness (marten), culvert height (weasel), through-culvert visibility (hare) and adjacent shrub cover (hare). A range of factors negatively affected culvert use by mammals (see paper for details). Thirty-six drainage culverts were monitored along a 55-km section of the Trans-Canada highway (two- and four-lane sections, with and without central reservation) and a 24-km section of highway 1A (two lanes, no central reservation). Crossings were determined from sooted track-plates (75 × 30 cm) in each culvert, checked weekly in January–April of 1999–2000 (≥ 12 times/culvert) and tracks in adjacent snow indicating culvert use.
A replicated study in 2000 along highways through two wetlands in British Columbia, Canada (Fitzgibbon 2001) found that culverts were used by small- to medium-sized mammals. Mammals used most of the eight dry culverts. In particular, there were frequent records of racoons Procyon lotor (on 11% of track plates) and species from the weasel family (on 32% of track plates – species not stated). Mice, voles and shrews combined were recorded on 31% of track plates. Racoons also used wet culverts on all nine occasions when tracks were not obscured by water. In 1995, twelve dry corrugated steel pipe culverts (average 35 long, 1 m diameter) were installed at 50-m intervals under a four-lane highway at one wetland. Eight were monitored. At another wetland, two wet cross-drainage corrugated steel pipe culverts (31 m long, 0.6 m diameter) were monitored. Aluminium track-plates, covered with soot, were installed 1–2 m inside each culvert and monitored over nine weekly intervals, in July–October 2000.
A review in 2000 of studies investigating whether mammalian predators use wildlife passages under roads and railways as ‘prey-traps’ (Little et al. 2002) found that most studies recorded no evidence of predation in or around passages. Evidence suggested that predator species used different passages to their prey. Only one study, in Australia, suggested that tunnels increased predation risk and that study recorded only one predator in tunnels. However, no studies specifically investigated predator activity, densities or predation rates, or predator-induced prey mortality at passage sites relative to control sites away from passages, or before-and-after passage construction. A literature survey was carried out in July 2000 using BIOSIS (Biological Abstracts) and Proceedings of the First, Second and Third International Conference on Wildlife Ecology and Transportation.
A study in 1998–1999 in a fragmented urban area in California, USA (Tigas et al. 2002) found that bobcats Felis rufus and coyotes Canis latrans used underpasses to cross a road. Nine road crossings (two by bobcats and seven by coyotes) out of 24 crossings where culverts were available within 100 m were through culverts and 15 (five by bobcats and 10 by coyotes) were over the road. Traffic levels were higher during crossings through culverts (2.1 cars/minute) than during crossings over the road (0.8 cars/minute). Results were not tested for statistical significance. The study was conducted northwest of Los Angeles from July 1998 to October 1999. Movements of 13 bobcats and nine coyotes were determined from 53 radio-tracking sessions (32 focussed on bobcats, 21 on coyotes). Locations were obtained every 30 minutes for 2–12 hours and road crossings were observed directly when possible.
A study in 2001–2003 along a highway through wetlands in Montana, USA (Foresman 2003) found that a range of mammals used culverts, including those with shelves fastened to sides. Twenty-three mammal species used culverts. These included six of the seven small mammal species that were recorded by trapping outside tunnel entrances; meadow vole Microtus pennsylvanicus, deer mouse Peromyscus maniculatus, vagrant shrew Sorex vagrans, Columbian ground squirrel Spermophilus columbianus, short-tailed weasel Mustela erminea and striped skunk Mephitis mephitis. Other mammals recorded using culverts included white-tailed deer Odocoileus virginianus, muskrat Ondatra zibethicus, raccoon Procyon lotor, coyote Canis latrans and red fox Vulpes vulpes. When water covered culvert floors, deer mice, short-tailed weasels, striped skunks and raccoons travelled along shelves in culverts. Meadow voles used tubes along culvert shelves. At least ten culverts (total number not clear) were monitored along a 6-mile section of Highway 93. Five had 25-inch-wide shelves installed. Culverts included some of 3–4 feet diameter and may have included others up to 10 feet wide. Monitoring was conducted from October 2001 to 2003 using heat- and motion-triggered cameras. Each month (March–October), small mammal populations adjacent to culverts were censused using 25 live traps, over three days.
A study in 2002 of mixed habitats including forest, swamp and farmland, along a highway in New York, USA (LaPoint et al. 2003) found that 19 culverts were rarely used as crossing points by mammals. The only crossings documented were five by northern racoons Procyon lotor at a single drainage culvert. Nineteen culverts were studied, along 141 km of highway, from 14 March to 29 April 2002. Culverts were categorised according to primary use: drainage (seven culverts), pedestrian underpass (nine), truck use (two) or bridge (one, where a river flowed beneath the road). Enabling white-tailed deer Odocoileus virginianus passage was also thought to be a motivation in installing at least some culverts. Animal passage was recorded using one camera trap at each culvert (average 40 days/site) and opportunistic snow-tracking when conditions permitted.
A replicated study in 1999–2000 along three major highways in California, USA (Ng et al. 2004) found that tunnels, culverts and underpasses were used by mammals. Fourteen of the 15 passages were used by racoons Procyon lotor (making 207 crossings), eight by opossums Didelphis virginianus (24 crossings), seven by coyotes Canis latrans (59 crossings), seven by bobcats Lynx rufus (36 crossings), five by striped skunks Mephitis mephitis (23 crossings), three by mule deer Odocoileus hemionus (26 crossings), one by spotted skunks Spilogale putorius (five crossings) and one by a mountain lion Puma concolor (one crossing). Crossing numbers include both verified and probable crossings. Rodents and cottontail rabbits Sylvilagus auduboni were also recorded. Six square livestock tunnels, five drainage culverts and four underpasses (surface roads or wide stream crossings) were studied. Passages were 44–218 m long and 2–238 m2 in cross-section. Camera traps were used in four passages and powder stations to detect animal footprints in 12 passages. One passage was monitored using both methods. Monitoring occurred over four consecutive days/month between July 1999 and June 2000.
A study in 2001–2003 on a road through rainforest in Queensland, Australia (Goosem et al. 2005) found that underpasses beneath the road were used by a range of mammals. There were 237 crossings recorded by brown bandicoots Isoodon obesulus, 233 by red-legged pademelons Thylogale stigmatica, 230 by coppery brushtail possums Trichosurus vulpecula johstoni, two by Lumholtz’s tree-kangaroos Dendrolagus lumholtzi, 53 by rodents and 13 by dogs Canis lupus familiaris or dingoes Canis dingo. Three underpasses (3.4 m high, 3.7 m wide), installed in 2001 below an upgraded two-lane road, were studied. Habitat enhancement features were added to each, such a soil, leaf and branch litter, rocks and logs and also vertical tree branches, to enable escape off the tunnel floor. Underpass use was monitored by weekly checks, over three years, for animal tracks in 1-m-wide strips of sand. Infrared-triggered cameras were used occasionally to confirm identifications.
A study in 2003 of a highway and railway in British Columbia, Canada (Krawchuk et al. 2005) found that at least two of three crossing structures were used by mammals. Mule deer Odocoileus hemionus were detected using one small culvert (2.1 m wide, 1.5 m high, 30 m long) six times. They were not recorded using a larger (7 m wide, 5 m high, 40 m long) cattle underpass though signs of their presence were noted nearby. Black bears were detected 20 times passing through the smaller culvert and four times through the cattle underpass. Raccoons were detected twice at the cattle underpass. The smaller culvert had a soil substrate, was surrounded by vegetation and was relatively far from human activity. The cattle underpass had limited surrounding natural vegetation. No mammals were recorded using a third culvert (1.2 m wide and high, 30 m long), possibly due to camera malfunction. Culverts and the underpass ran under both the Trans-Canada Highway and Canadian Pacific Railway. They were monitored using infrared sensor cameras during August–November 2003. Animal tracks or signs around camera stations were also recorded.
A study in 2004–2006 in an area of rice fields and scattered forest in Jeollanamdo province, South Korea (Choi & Park 2007) found that highway underpasses were used by a range of mammals, though road sections with higher underpass density did not have fewer wildlife-vehicle collisions. Eleven wild mammal species were recorded using underpasses. The most frequent were raccoon dog Nyctereutes procyonoides (865 images), brown rat Rattus norvegicus (455), leopard cat Prionailurus benalensis (253), striped field mouse Apodemus agrarius (229), Siberian weasel Mustela sibirica (166), Eurasian otter Lutra lutra (35) and water deer Hydropotes inermis (32). Ninety-three roadkill mammals of 12 species were recorded. The most frequent were rodents (24 casualties), leopard cat (17), Siberian weasel (13) and water deer (12). Most mammals used all underpass types frequently, except water deer, which rarely used small passages. Use of seven circular culverts (0.8–1.2 m diameter), two box culverts (2.5 m wide and high) and five human underpasses (2.0–4.3 m wide and high), selected from 31 underpasses along a 6.6-km section of four-lane highway, were monitored from September 2005–August 2006. One or two infrared-operated cameras were installed 1–2 m inside each underpass for an average of 239 days/underpass. Wildlife-vehicle collisions were recorded daily from September 2004–August 2006.
A study in 2004–2005 at seven sites along roads through forest in Virginia, USA (Donaldson 2007) found that white-tailed deer Odocoileus virginianus used underpasses to cross the road but black bears Ursus americanus did not. White-tailed deer crossed through four of seven underpasses monitored, with a total of 1,107 crossings detected. Black bears approached one underpass entrance three times, but did not cross through. Other mammals recorded in underpasses included opossums Didelphis virginiana, bobcats Lynx rufus, red foxes Vulpes vulpes, coyotes Canis latrans, raccoons Procyon lotor and groundhogs Marmota monax as well as squirrels and mice (see paper for details). Seven underpasses were monitored. Five were culverts (1.8–6.1 m wide, 1.8–4.6 m high and 21–79 m long). Two were crossings under bridges (13–94 m wide, 5–14 m high and 10–18 m long). Underpasses were not fenced and most had a narrow water section. Underpasses were monitored from June 2004 to May 2005, using one or two camera traps at each entrance.
A study in 2003–2005 along a highway through deciduous woodland in North Carolina, USA (Kleist et al. 2007) found that mammals used a wildlife underpass. An estimated 299 mammal crossings of at least 10 species occurred (based on 126 crossings observed on a sample of video surveillance). Of these, an estimated 185 were white-tailed deer Odocoileus virginianus crossings. At least 17 deer approached the underpass but retreated without crossing. Other mammals crossing included red or gray fox Vulpes vulpes or Urocyon cinereoargenteus, raccoon Procyon lotor, woodchuck Marmota monax, gray squirrel Sciurus carolinensis and chipmunk Tamias striatus. Only four incidences of mammals killed by vehicles were recorded from December 2003 to June 2005. Two digital ultra-low-light video cameras and infrared spotlights monitored underpass use below a four-lane highway between December 2003 and May 2005. A sample of videos was viewed from 458 days of continual video recordings. The underpass was constructed in 1955, encompassing a 6-m width either side of a stream. It was 2–3 m high and 41 m long. Weekly surveys of vehicle-killed animals were undertaken on a 1.8-km section of road encompassing the underpass.
A global review in 2007 of 123 studies investigating the use of 1,864 wildlife crossings (van der Ree 2007) found that all studies reported that the majority of underpasses and overpasses were used by wildlife. Of the 1,864 structures reported on, most were underpasses (83%), including culverts (742 examples), bridges (130), tunnels (340) and unknown types (333). 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 review of 30 studies reporting on monitoring of 329 crossing structures in Australia, Europe and North America (Taylor & Goldingay 2010) found that mammals used most culverts and underpasses. Small mammals used pipes (demonstrated by 6/7 relevant studies), drainage culverts (5/5 studies), adapted culverts (5/5 studies), wildlife underpasses (3/4 studies) and bridge underpasses (2/3 studies). Arboreal mammals used pipes (1/1 studies), drainage culverts (4/4 studies), adapted culverts (4/4 studies) and bridge underpasses (1/1 studies). Medium-sized mammals used pipes (8/11 studies), drainage culverts (12/13 studies), adapted culverts (8/8 studies), wildlife underpasses (6/8 studies) and bridge underpasses (6/7 studies). Large mammals used pipes (6/9 studies), drainage culverts (11/12 studies), adapted culverts (11/11 studies), wildlife underpasses (24/24 studies) and bridge underpasses (14/15 studies). Larger mammals tended to use more open underpasses. Small and medium-sized mammals used underpasses with funnel-fencing or adjoining walls and those with vegetation cover close to entrances. Those with vegetation cover tended to be avoided by some ungulates. Thirty papers reporting monitoring of 329 crossing structures were reviewed. Fourteen papers investigated multiple structure types, resulting in a total of 52 studies of different structure types. Underpasses, from small drainage pipes to dry passage bridges, comprised 82% of crossings.
A study in 2010 of a road through forest and pastureland in New South Wales, Australia (Crook et al. 2013) found that bare-nosed wombats Vombatus ursinus used culverts to cross the road. Bare-nosed wombats used eight out of 19 monitored culverts. Wombats were recorded using culverts on 16 out of 190 camera-trap nights. One culvert was used three times in one night and three were used twice in one night. Other culverts were not used more than once in a night. The study was conducted along 8 km of a two-lane road. Nineteen concrete pipe culverts (40–60 cm diameter and 13–25 m long) were monitored between April and August 2010. A camera trap was set 1 m from each culvert entrance for 10 days. Five culverts were dry with earth substrate, nine were dry without earth substrate and five had constant water flow. Culverts were 40–2,200 m apart.
A study in 2009 at 10 sites along a highway through forest in Alberta, Canada (D'Amico et al. 2015) found that North American deer mice Peromyscus maniculatus used underpasses to cross a road but meadow voles Microtus pennsylvanicus and southern red-backed voles Myodes gapperi did not. Tracks of deer mice were recorded in 90% of track tubes in elliptical culverts, in 87% of track tubes in box culverts and in 75% of track tubes on open-span bridge underpasses. No tracks of meadow vole or southern red-backed vole were detected, despite their use of overpasses in the area. Over two weeks in September–October 2010, small mammals were surveyed in three elliptical metal culverts (4 m high, 7 m wide), five concrete box culverts (2.6 m high, 3.2 m wide) and two open-span bridge underpasses (3 m high, 11 m wide). Underpasses were unvegetated and entrances were characterized by roadside grasslands. Two parallel sample lines, each of five 30 × 10 cm track tubes with sooted metal sheet as a floor, were placed in the centre of each underpass. Mammals were identified from their footprints.
A study in 2015 along a highway in Montana, USA (Andis et al. 2017) found that underpasses were used by white-tailed deer Odocoileus virginianus and mule deer Odocoileus hemionus more than expected compared to their movements through adjacent habitats, but no difference was found for black bear Ursus americanus or coyote Canis latrans. Overall, white-tailed deer (recorded at all 15 underpasses) and mule deer (at five of 15 underpasses) had an average of 88% and 472% more movements/day respectively through underpasses than adjacent habitats. Black bear (recorded at seven of 15 underpasses) and coyote (at 13 of 15 underpasses) had an average of 112% and 75% more movements/day respectively through underpasses than adjacent habitats, but the difference was not significant. Fifteen elliptical underpasses were installed in 2006–2011 along a 91 km stretch of highway. Underpasses (7–8 m wide, 4–6 m high, 15–40 m long) were constructed from corrugated metal with a soil substrate and retaining walls extending 10 m from the roadside. Twelve of the 15 underpasses had 2.4-m high wildlife exclusion fencing. Infrared cameras recorded large mammal movements through each underpass (one camera/entrance) and at random locations within an adjacent 300 m2 plot on each side (five cameras/plot) for 12–20 days in April–November 2015.
A replicated study in 2008–2011 of 265 culverts throughout Maryland, USA (Sparks & Gates 2017) found that culverts were used by a range of mammal species to cross roads. Crossings were made by northern raccoons Procyon lotor (0.79/culvert/day), Virginia opossums Didelphis virginiana (0.03/culvert/day), woodchucks Marmota monax (0.03/culvert/day), red foxes Vulpes vulpes (0.03/culvert/day), eastern gray squirrels Sciurus carolinensis (0.02/culvert/day) and both common gray foxes Urocyon cinereoargenteus and white-footed mice Peromyscus spp (0.01/culvert/day). Between August 2008 and January 2011, a total of 265 randomly selected culverts were monitored using camera traps for a total of 31,317 camera-trap days. Culverts were located under paved roads and contained either a waterway, a route for water flow, or other depression. Culverts averaged 2.4 m wide, 1.9 m high and 46.4 m long. Each culvert was sampled at least nine times in 2008–2011, for 10–36 days each time, using one camera trap. The camera was placed at the approximate midpoint of the culvert or near the entrance.
- Mansergh I.M. & Scotts D.J. (1989) Habitat continuity and social organisation of the mountain pygmy-possum restored by tunnel. The Journal of Wildlife Management, 53, 701-707
- Woelfel H. & Krueger H.H. (1995) On the design of game passages across highways. Zeitschrift für Jagdwissenschaft, 41, 209-216
- Yanes M., Velasco J.M. & Suarez F. (1995) Permeability of roads and railways to vertebrates: the importance of culverts. Biological Conservation, 71, 217-222
- Rosell C., Parpal J., Campeny R., Jove S., Pasquina A. & Velasco J.M. (1997) Mitigation of barrier effect on linear infrastructures on wildlife. Pages 367-372 in: Habitat Fragmentation & Infrastructure. Ministry of Transport, Public Works and Water Management, Delft, Netherlands.
- Norman T., Finegan A. & Lean B. (1998) The role of fauna underpasses in New South Wales. Proceedings of the 1998 International Conference on Wildlife Ecology and Transportation, Florida Department of Transportation, Tallahassee, Florida USA, 195-208.
- Veenbaas G. & Brandjes J. (1999) Use of fauna passages along waterways under highways. Proceedings of the International Conference on Wildlife Ecology and Transportation, Florida Department of Transportation, Tallahassee, Florida USA, 253-258.
- Austin J.M. & Garland L. (2001) Evaluation of a wildlife underpass on Vermont State Highway 289 in Essex, Vermont. Proceedings of the 2001 International Conference on Ecology and Transportation, Center for Transportation and the Environment, North Carolina State University, Raleigh NC, USA, 616-624.
- Clevenger A.P., Chruszcz B. & Gunson K. (2001) Drainage culverts as habitat linkages and factors affecting passage by mammals. Journal of Applied Ecology, 38, 1340–1349
- Fitzgibbon K. (2001) An evaluation of corrugated steel culverts as transit corridors for amphibians and small mammals at two Vancouver Island wetlands and comparative culvert trials. MA thesis. Royal Roads University, Vancouver, Canada.
- Little S.J., Harcourt R.G. & Clevenger A.P. (2002) Do wildlife passages act as prey-traps? Biological Conservation, 107, 135-145
- Tigas L.A., Van Vuren D.H. & Sauvajot R.M. (2002) Behavioral responses of bobcats and coyotes to habitat fragmentation and corridors in an urban environment. Biological Conservation, 108, 299-306
- Foresman K.R. (2003) Small mammal use of modified culverts on the Lolo South project of western Montana – an update. Proceedings of the 2003 International Conference on Ecology and Transportation, Center for Transportation and the Environment, North Carolina State University Raleigh NC, USA, 342-343.
- LaPoint S., Keys R.W. & Ray J.C. (2003) Animals crossing the Northway: are existing culverts useful? Adirondack Journal of Environmental Studies, 10, 11-17
- Ng S.J., Dole J.W., Sauvajot R.M., Riley S.P.D. & Valone T.J. (2004) Use of highway undercrossings by wildlife in southern California. Biological Conservation, 115, 499-507
- Goosem M., Weston N. & Bushnell S. (2005) Effectiveness of rope bridge arboreal overpasses and faunal underpasses in providing connectivity for rainforest fauna. Proceedings of the 2005 International Conference on Ecology and Transportation, Center for Transportation and the Environment, North Carolina State University, Raleigh NC, USA, 304-318.
- Krawchuk A., Larsen K.W., Weir R.D. & Davis H. (2005) Passage through a small drainage culvert by mule deer, Odocoilus hemionus, and other mammals. The Canadian Field-Naturalist, 119, 296-298
- Choi T. & Park C.H. (2007) Can wildlife vehicle collision be decreased by increasing the number of wildlife passages in Korea? Proceedings of the 2007 International Conference on Ecology and Transportation, Center for Transportation and the Environment, North Carolina State University, Raleigh NC, USA, 392-400.
- Donaldson B. (2007) Use of highway underpasses by large mammals and other wildlife in Virginia: factors influencing their effectiveness. Transportation Research Record, 2011, 157-164
- Kleist A.M., Lancia R.A. & Doerr P.D. (2007) Using video surveillance to estimate wildlife use of a highway underpass. The Journal of Wildlife Management, 71, 2792-2800
- van der Ree R., van der Grift E., Mata C. & Suàrez F. (2007) Overcoming the barrier effect of roads –how effective are mitigation strategies? An international review of the use and effectiveness of underpasses and overpasses designed to increase the permeability of roads for wildlife. Proceedings of the 2007 International Conference on Ecology and Transportation, Center for Transportation and the Environment, North Carolina State University, Raleigh NC, USA, 423-431.
- Taylor B.D. & Goldingay R.L. (2010) Roads and wildlife: impacts, mitigation and implications for wildlife management in Australia. Wildlife Research, 37, 320-331
- Crook N., Cairns S.C. & Vernes K. (2013) Bare-nosed wombats (Vombatus ursinus) use drainage culverts to cross roads. Australian Mammalogy, 35, 23-29
- D'Amico M., Clevenger A.P., Román J. & Revilla E. (2015) General versus specific surveys: Estimating the suitability of different road-crossing structures for small mammals. The Journal of Wildlife Management, 79, 854-860
- Andis A.Z., Huijser M.P. & Broberg L. (2017) Performance of arch-style road crossing structures from relative movement rates of large mammals. Frontiers in Ecology and Evolution, 5, 122
- Sparks J.L. & Gates J.E. (2017) Seasonal and regional animal use of drainage structures to cross under roadways. Human Wildlife Interactions, 11, 182-191