Release captive-bred individuals into the wild to restore or augment wild populations of wildfowl
Overall effectiveness category Unknown effectiveness (limited evidence)
Number of studies: 5
Background information and definitions
Captive breeding is normally used to provide individuals which can then be released into the wild to either restore a population in part of the speciesâ€™ former range, or to augment an existing population.
Release techniques vary considerably, from â€˜hard releasesâ€™ involving the simple release of individuals into the wild to â€˜soft releasesâ€™ which involve a variety of adaptation and acclimatisation techniques before release or post-release feeding and care. The following section includes studies describing the overall effects of release projects. Studies that compare specific release techniques are described elsewhere (â€˜Use holding pens at release sitesâ€™, â€˜Use â€˜anti-predator trainingâ€™ to improve survival after releaseâ€™ etc).
Supporting evidence from individual studies
A 1997 review of the Hawaiian goose (nene) Branta sandvicensis reintroduction programme (Black et al. 1997) concluded that the release of 2,150 captive-bred birds on Hawaii and Maui, USA, starting in 1949 had not resulted in a self-sustaining wild population. Estimated mortality rates ranged from 0-87% annually, although were generally low until droughts in 1973-86, when 1,200 released geese died. Mortality rates were lower in the lowest-altitude release site (at <1,300 m a.s.l.), with only three years between 1976 and 1983 having mortality rates over 15%. By contrast, the few geese released in uplands that survived the droughts did so by migrating away from their release site. Over the study period there were 515 nests recorded, with 37% raising at least one gosling. Overall there were 473 goslings raised (0.92 goslings/nest), with the highest rates in lowland sites. Birds were all released into temporary enclosure, with differences in release techniques discussed in â€˜Clip birdsâ€™ wings on releaseâ€™ and â€˜Release birds as adults or sub-adults, not juvenilesâ€™.Study and other actions tested
A replicated, controlled study in wetlands in southwest Manitoba, Canada, between 1992 and 1995 (Yerkes & Bluhm 1998) found that only 2-3% of 1,766 released captive-bred female mallards Anas platyrhynchos were re-sited close to the release site (36 females positively identified and 19 more possibly identified). Annual rates ranged from nearly 10% of 1992 releases to only 1% of 1994 releases being seen again. If these numbers are adjusted for an average mortality of 60% for juvenile females, return rates were still only 6-9%. Comparisons of reproductive success were difficult due to small sample sizes (12 captive-reared females and 30 wild females were monitored): with 60% of wild females and either 71% (1993) or 0% (1994) of captive-reared females nesting. Nest success was 80% for five captive-reared females; it ranged from 11% (1993) to 67% (1994) for wild females, dependent on whether the majority nested on the ground or on artificial structures. Releases involved acclimatising groups of ducklings for 6-10 hours in open-topped cages before allowing them out. Supplementary food was also provided for three weeks after release (see â€˜Provide supplementary after releaseâ€™ for more information on this release technique).Study and other actions tested
A before-and-after study from the Aleutian Islands, Alaska, USA on the cackling goose Branta hutchinsii recovery programme (USFWS 2001) found that the goose population increased from fewer than 1,000 birds in the 1970s to over 6,000 by 1991, following the release of captive-bred birds and the eradication of Arctic foxes Alopex lagopus from breeding islands (see â€˜Predator control on islandsâ€™). The authors note, however, that the release of captive-bred geese was not very successful overall. Release techniques are discussed in â€˜Release birds as adults or sub-adults, not juvenilesâ€™ and â€˜Clip birdsâ€™ wings on releaseâ€™.Study and other actions tested
A replicated study at a reintroduction programme in the north of North Island, New Zealand (O\'Connor 2005) found that 60 captive-bred brown teal Anas chlorotis released in 2003 had an annual survival rate of just 45%, but the survival rate of 40 individuals released in 2004 was 85%. This difference was probably due to a more intensive control of feral cats Felis catus (a major cause of mortality in 2003) between releases (see â€˜Control predators not on islandsâ€™). The site also had on-going control of stoats Mustela erminea and ferrets M. putorius. The authors suggest that higher survival may also have been due to the presence of an established teal population in 2004 but not 2003. Teal were not kept in aviaries at the site before release, but were provided with supplementary food (see â€˜Provide supplementary food after releaseâ€™ and â€˜Use holding pens at release sitesâ€™ for details on these techniques).Study and other actions tested
A reintroduction programme on Campbell Island, New Zealand, in 2004-5 (McClelland & Gummer 2006) found that at least 78% (2004) and 75% (2005) of 105 Campbell Island teal Anas nesiotis survived reintroduction or translocation. The birds also bred in 2006, with at least two nests and four young being produced. Forty-four of the released birds were wild-caught (see â€˜Translocate individualsâ€™) and 61 captive bred. All birds were kept individually or in pairs for 2-10 days in small holding pens on Campbell Island and provided with food before being released into the wild.Study and other actions tested