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: Remove or control predators Terrestrial Mammal Conservation

Key messages

Read our guidance on Key messages before continuing

  • Ten studies evaluated the effects on non-controlled mammals of removing or controlling predators. Seven studies were in North America, one was in Finland, one in Portugal and one in Mexico.



  • Abundance (6 studies): Three of six studies (including three controlled, one before-and-after and one replicated, paired sites study), in Finland Portugal, Mexico and the USA, found that removing predators increased abundances of pronghorns, moose and European rabbits and Iberian hares. One of these studies also found that mule deer abundance did not increase. The other three studies found that removing predators did not increase mountain hare, caribou or desert bighorn sheep abundance.
  • Reproductive success (2 studies): Two replicated, before-and-after studies (one also controlled), in the USA, found that predator removal was associated with increased breeding productivity of white-tailed deer and less of a productivity decline in pronghorns. However, one of these studies also found that there was no change in breeding productivity of mule deer.
  • Survival (5 studies): Two of five before-and-after studies (including two controlled studies and one replicated study), in the USA, Canada and the USA and Canada combined, found that controlling predators did not increase survival of caribou calves, or of calf or adult female caribou. Two studies found that moose calf survival and woodland caribou calf survival increased with predator control. The other study found mixed results with increases in white-tailed deer calf survival in some but not all years with predator control.


Supporting evidence from individual studies


A replicated, paired sites, controlled study in 1993–1998 of boreal forest in three areas in Finland (Kauhala et al. 1999) found that removing predators did not increase numbers of mountain hares Lepus timidus. In two of three areas, mountain hare numbers increased in both predator removal and predator protection sites, with the rate of increase being higher in the predator protection site than the removal site in one of those areas. In the third area, hare numbers declined each year in predator removal sites but increased in two of five years in protection sites. Data are presented as track count indices. In each of three areas, a predator removal and predator protection site were established, ≥5 km apart. Sites each covered 48–116 km2. Predator removal, carried out by hunters during normal hunting seasons, commenced in August 1993, targeting red fox Vulpes vulpes, pine marten Martes martes, stoat Mustela ermine and raccoon dog Nyctereutes procyonoides. Hares were monitored by snow track counts, annually from 15 January to 15 March, in 1993–1998.


A controlled, before-and-after study in 1990–2000 in alpine tundra and subalpine shrubland in Alaska, USA (Valkenburg et al. 2004) found that wolf Canis lupus culling did not increase calf survival or population size of caribou Rangifer tarandus. Between 1992-1993 (before the wolf cull) and 1994-1995 (after the cull), the increase in calf:cow ratio within the cull area (before: 7.4:100; after: 21.5:100) was no greater than in a similar sized herd in an area without wolf culling (before: 11.2:100; after: 19.5:100). However, the change was greater than in a smaller sized herd in an area without wolf culling, where the calf:cow ratio declined (before: 15.8:100; after: 11.5:100). The long-term (1993–2000) change in caribou numbers in the population where wolves were controlled (before: 3,661; after: 3,227) was comparable to the population change in one of the areas without culling (before: 1,970; after: 1,730), but not to the other (before: 500; after: 675), although no statistical tests were carried out. Autumn calf:cow ratios were monitored annually between 1990 and 2000 from a helicopter, guided by radio-collared females. See original paper for methods for estimating population size. In 1993–1994, 60–62% of wolves were controlled by trapping, snaring and shooting. Smaller numbers (20–40%) were culled in subsequent years by local hunters.


A replicated, paired sites study in 2000–2001 of 24 games estates and hunting areas in Alentejo, Portugal (Beja et al. 2009) found that controlling predators resulted in greater numbers of European rabbits Oryctolagus cuniculus and Iberian hares Lepus granatensis. Game estates that controlled predators had a greater number of European rabbits (5.9 rabbits/10 km) and Iberian hares (1.7 hares/10 km) than paired hunting areas without predator control (0.5 rabbits/10 km; 0.3 hares/10 km). Twelve game estates that controlled predators (with box traps, shooting, snares) for >3 years were paired with 12 hunting areas without predator control. Paired sites (average 12 km2) were mostly grazed woodlands and farmland. Species controlled were red foxes Vulpes Vulpes (11 estates), Egyptian mongooses Herpestes ichneumon (six estates), feral cats Felis catus and dogs Canis familiaris (two estates), common genets Genetta genetta (one estate), stone martens Martes foina (one estate) and azure-winged magpies Cyanopica cyanus (one estate). Each site within a pair was sampled once on consecutive days in May–June 2000 or 2001. Rabbits and hares and/or their signs (faeces, footprints) were counted along walked transects (average 12 km long).


A replicated study in 1951–2007 in nine desert sites in Arizona and New Mexico, USA, and the Gulf of California, Mexico (Wakeling et al. 2009) found that controlling mountain lions Puma concolor did not increase the population size of desert bighorn sheep Ovis canadensis. No bighorn sheep populations at sites where mountain lions were controlled increased in size (data not presented). Data were obtained from historical records for 10 sites with long-term survey and hunting information. Data included counts of bighorn sheep from both surveys and hunter harvests, and annual mountain lion harvests. No information on the number of mountain lions controlled is provided.


A replicated, paired sites, controlled, before-and-after study in 2007–2008 in 12 rangeland sites in Wyoming and Utah, USA (Brown and Conover 2011) found that after coyotes Canis latrans were removed, pronghorn Antilocapra americana abundance was higher and productivity declined less in removal than non-removal sites, but for mule deer Odocoileus hemionus abundance and productivity did not differ. After eight months of coyote control, the abundance of pronghorn was higher and decline in productivity smaller in removal (abundance: 4.4 pronghorn/km2; change in productivity: –6.5 fawns/100 adult females) than in non-removal sites (abundance: 2.5 pronghorn/km2; change in productivity: –22 fawns/100 adult females). However, mule deer abundance and productivity did not differ between removal (abundance: 3.5 mule deer/km2; productivity: 56 fawns/100 adult females) and non-removal sites (abundance: 4.9 mule deer/km2; productivity: 62 fawns/100 adult females). Six pairs of sites in similar habitat were selected. Site areas totalled 10,517 km2. Between late July 2007 and March 2008, an average of 195 coyotes/1,000 km2 were removed from one site in each pair by trapping and shooting. Pronghorn and mule deer were counted by driving 17–27 km-long transects at 25 km/hr weekly during July and August and fortnightly in September, in 2007 and 2008.


A before-and-after study in 2001–2007 in a mosaic of shrub, forest and taiga in Alaska, USA (Keech et al. 2011) found that control of American black bear Ursus americanus, brown bear Ursus arctus and wolf Canis lupus increased moose Alces alces abundance and calf survival. Moose abundance and calf survival were higher after predator control (abundance: 0.56 moose/km2; calf/adult ratio: 51–63 calves/100 adult females) than before control (abundance: 0.38 moose/km2; calf/adult ratio: 34 calves/100 adult females). In May 2003 and 2004, 109 black and nine brown bears were translocated at least 240 km from a 1,368-km2 area, reducing the populations by approximately 96% and 50% respectively. In 200–2008, wolf numbers were reduced by 11–33 animals/year across a wider 8,314-km2 area by aircraft-assisted shooting, conventional hunting and trapping (density in 2001: 5.1 wolves/1,000 km2; density in 2006: 1.3 wolves/1,000 km2). Aircraft surveys (3.1 min/km2) were used to monitor moose numbers and calf/adult ratios annually, in autumn, at 87 sites within the study area, each of 15.7 km2.


A replicated, before-and-after study in 2006–2012 in three forest sites in South Carolina, USA (Kilgo et al. 2014) found that control of coyotes Canis latrans increased fawn survival in white-tailed deer Odocoileus virginianus in two out of three years. The annual survival rate of deer calves was higher one year (0.51) and three years (0.43) after the start of coyote control than before control (0.23), but did not differ two years (0.20) after the start of coyote control. The percentage of fawn mortalities that resulted from predation by coyotes was similar after (73%) compared to before control (80%). Between mid-January and early April 2010–2012, four hundred and seventy-four coyotes were removed from three 32-km2 sites (1.6 coyotes /km2/year) by trapping. The survival of 216 fawns (91 before and 125 after coyote control) was monitored using motion-sensitive radio-collars. Calves were monitored every eight hours if younger than four weeks, 1–3 times/day up to 12 weeks of age, weekly up to 16 weeks and 1–4 times/month up to 12 months.


A replicated, before-and-after study in 2010–2013 in two forest sites in Georgia, USA (Gulsby et al. 2015) found that controlling coyotes Canis latrans increased the number of young white-tailed deer Odocoileus virginianus relative to adult females in one of two sites. In one of two sites the number of young white-tailed deer was higher after coyote control (1.01 fawns/adult female) compared to before control (0.63 fawns/adult female). However, in one site there was no significant difference (after control: 0.85 fawns/adult female; before control: 0.84 fawns/adult female). Coyote abundance was lower after control (4–16 animals/site) than before control (16–21 animals/site). In March–June 2011, professional trappers controlled coyotes in both sites. In January and February of 2010–2013, infrared cameras were arranged in a grid pattern, over a 2,000-ha area, at a density of 1 camera/65 ha at each site. Cameras were baited with corn and took a photograph every 15 minutes for 10 days. The number of pictures of young deer relative to pictures of adult females was calculated.


A before-and-after study in 1994–2002 in a large forest and shrubland area in Alaska, USA and Yukon, Canada (Boertje et al. 2017) found that trapping and removing or sterilizing wolves Canis lupus did not reduce caribou Rangifer tarandus mortality. The annual mortality of caribou calves (≤1 year old) did not differ after wolf removal or sterilization commenced (50–67%) compared to before (39–65%). Adult female (≥1 year old) annual mortality was also similar after wolf removal or sterilization commenced (9–10%) compared to before (9%). In a 50,000-km2 study area, 52–78 newborn caribou calves/year were radio-collared in May 1994–2002. Caribou were monitored during ≥3 flights/year. In 15 wolf packs, the dominant pair was sterilized in November 1997 and remaining wolves in those packs were translocated, mainly in April 1998. Eight additional packs were similarly treated over the following two winters. Caribou mortality was measured over four years before and five years after wolf control commenced.


A controlled, before-and-after study in 2008–2013 in four boreal forest, peatland and heath sites in Newfoundland, Canada (Lewis et al. 2017) found that controlling coyotes Canis latrans increased caribou Rangifer tarandus calf survival. Caribou calf survival was higher when coyotes were controlled (70-day survival: 41%; 182-day survival: 32%) compared to before coyote control was carried out (70-day survival: 9%; 182-day survival: 7%). Survival rates across these two periods at sites without coyote control were stable (70-day survival: 52–58%; 182-day survival: 47%). At one site (covering 480 km2), lethal neck snares were set in March or April of 2012 and 2013 and were removed one week before caribou calving commenced in May. Forty coyotes were removed over these two years. Coyotes were not controlled at three other caribou calving sites. Caribou calves were radio-collared in late May to early June of 2008–2009 (193 calves) and 2012–2013 (103 calves), when 1–5-days old, and were monitored by radio-tracking through to November.

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.