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 fences around existing culverts or underpasses under roads/railways Terrestrial Mammal Conservation

Key messages

Read our guidance on Key messages before continuing

  • Four studies evaluated the effects on mammals of installing fences around existing culverts under roads/railways. Two studies were in the USA one was in Portugal and one was in South Africa.



  • Survival (3 studies): Two out of three before-and-after studies (including a controlled and a site comparison study), in the USA, Portugal and South Africa, found that installing or enhancing roadside fencing alongside existing culverts reduced mammal road mortality whilst one study found that such fences did not alter mammal road mortality.


Supporting evidence from individual studies


A before-and-after study in 1976–1981 along a highway through shrubland in Wyoming, USA (Ward 1982) found that after a fence alongside the highway that was connected to underpasses was made taller, fewer mule deer Odocoileus hemionus were killed. Results were not tested for statistical significance. In six migration seasons (three springs, three autumn–winters) after increasing the height of the fence, only one deer-vehicle accident occurred in the fenced area. In three migration seasons before fence construction (two spring and one autumn–winter), 53 deer–vehicle accidents occurred within the area to be fenced. The study was conducted along a stretch of highway constructed in late 1970. In 1977–1978, the height of a fence along the highway was increased from 4 ft to 8 ft along both sides of 7.8 miles of road. The fence allowed deer to access seven underpasses (length: 110–393 feet; width: 10–50 feet; height: 10–17 feet). Deer movement was monitored before (1976–1977) and after (1978–1981) fence heightening by direct observation, track counts, radio-tracking and automatic cameras. The highway was located across a migration route of 1,600–2,000 mule deer.


A replicated, randomized, controlled, before-and-after study in 1997–1999 in dry shrubland along a highway in Texas, USA (Cain et al. 2003) found that installing fences to funnel animals to existing culvert entrances did not increase culvert use by bobcats Lynx rufus. Fences did not significantly increase cat spp. use of culverts (data not presented). However, among four culverts most used by bobcats, two fenced culverts saw a rise in use after fence installation (after 7.2; before: 3.9 track sets/month) while two unfenced culverts saw a fall over this same time (after: 2.2; before: 2.9 track sets/month). Most cats (371 of 471 camera-trap images) were bobcats. The remainder were feral cats Felis catus. At six culverts, randomly selected from 12, wire net fences (1.6 m high) were erected at entrances, extending 100 m to each side, parallel to the road. Culverts were checked two times/week from 1 July 1997 to 31 May 1999 for cat spp. tracks. Remote cameras were used at culverts from 1 August 1997 to 31 May 1999. Fences were erected after the first year of monitoring.


A replicated, before-and-after, site comparison study in 2008–2009 of 64 culverts under roads in southern Portugal (Villalva et al. 2013) found that fences connecting to existing under-road culverts did not alter mammal road mortality. After fence installation, there was a similar number of mammals killed by traffic (19 road-kills) compared to before (20 road-kills). There was also no significant difference in mammal road-kills between road sections where fences were installed (19 road-kills) and those that were not fenced (13 road-kills). In April 2008, 100-m-long fences with 2.5-cm mesh, buried to 50 cm deep and extending 50 cm above ground, were installed alongside the road at each side of 32 under-road culverts. These were in addition to existing livestock fencing. Another 32 culverts in the same area that were unfenced were selected for comparison. The number of mammals killed by traffic was recorded by highway maintenance staff for 10 months before and 10 months after fence installation.


A randomized, replicated, controlled, before-and-after study in 2015 along a road through dry savanna in Limpopo, South Africa (Collinson et al. 2017) found that installing fences around existing culverts reduced mammal road casualties. Results were not tested for statistical significance. One scrub hare Lepus saxatilis was detected as a roadkill near fenced culverts compared to two bushveld gerbils Tatera leucogaster detected as roadkills before fencing was installed. Concurrently, two multimammate rats Mastomys sp. were detected as roadkills near unfenced culverts after fence installation at treatment sites compared to one before fence installation. The study was conducted along six 400-m-long road segments with culverts. In three segments, a 70-cm-high fence was erected extended 200 m along both sides of the road on either side of the culvert. The fence was approximately 2 m from the road verge, sloped at 45° away from the road and extended 30 cm below ground. Three segments remained unfenced. Roadkills were counted in all sites during a 20-day period before fences were installed (January 2015) and a 20-day period after (February 2015). Roadkills were counted by an observer in a car moving at 40–50 km/h.

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.