Reedbeds as refuges for water voles (Arvicola terrestris) from predation by introduced mink (Mustela vison)

The water vole has recently undergone a catastrophic decline in the UK. This has resulted from loss and fragmentation of suitable habitat leaving water vole populations highly vulnerable to the impact of predation by introduced American mink. However, at some reedbed sites water voles and mink have apparently coexisted for many years. To determine if reedbeds offer a refuge from predation, 70 voles were radio-tagged at three sites in England and overwinter mortality monitored. Water vole perception of predation risk was also assessed. Mortality was high (64%), predation by mustelids, including mink, being the chief cause. Experiments suggested that voles failed to perceive areas of highest predation risk. However, predation rate declined strongly with the distance water voles lived from a main water channel. Thus, reedbeds provide a refuge from predation, even by mink, and calculations suggest that they may support source populations enhancing the viability of water vole metapopulations. Consequently reedbeds are now being used as one focus for the conservation of water voles in England and Wales.     

Vulnerability of ground-nesting waterbirds to predation by invasive American mink in the Cape Horn Biosphere Reserve, Chile

Biological invasions constitute one of the most important threats to biodiversity. This is especially true for “naïve” birds that have evolved in the absence of terrestrial predators in island ecosystems. The American mink (Mustela vison) has recently established a feral population on Navarino Island (55°S), southern Chile, where it represents a new guild of terrestrial mammal predators. We investigated the impact of mink on ground-nesting coastal waterbirds with the aim of deriving a vulnerability profile for birds as a function of different breeding strategies, habitat, and nest characteristics. We compared rates of nest survival and mink predation on 102 nests of solitary nesting species (Chloephaga picta, Tachyeres pteneres), on 361 nests of colonial birds (Larus dominicanus, Larus scoresbii, Sterna hirundinacea), and on 558 artificial nests. We calculated relative mink and bird densities at all nest sites. Nests of colonial species showed the highest nest survival probabilities (67-84%) and no predation by mink. Nest survival rates for solitary nesting species were lower (5-20%) and mink predation rates higher (10-44%). Discriminant analyses revealed that mink preyed upon artificial nests mainly at shores with rocky outcroppings where mink were abundant. High nest concealment increased the probability for predation by mink. Conservation planning should consider that invasive mink might severely affect the reproduction success of bird species with the following characteristics: solitary nesting, nesting habitat at rocky outcrop shores, and concealed nests. We recommend that work starts immediately to control the mink population with a priority in the nesting habitats of vulnerable endemic waterbirds.     

The effects of captive experience on reintroduction survival in carnivores: A review and analysis

This review focuses on the success and survivorship of captive-born versus wild-caught carnivores used in reintroductions. Previous reviews have suggested that reintroduction projects using captive-born animals are less likely to be successful than projects translocating wild-caught animals. The purpose of this paper is to examine this statistically and investigate how captivity may affect the survival of reintroduced carnivores. We examined results published in previous reviews, and found evidence to support that reintroduction projects using wild-caught animals are significantly more likely to succeed than projects using captive-born animals. We further compiled our own review of 45 case studies in carnivore reintroduction projects (in 17 species across 5 families) to investigate survival rates rather than overall project ‘success’. We found that (1) wild-caught carnivores are significantly more likely to survive than captive-born carnivores in reintroductions; (2) that humans were the direct cause of death in over 50% of all fatalities and (3) that reintroduced captive-born carnivores are particularly susceptible to starvation, unsuccessful predator/competitor avoidance and disease.     

Trapping for mink control and water vole survival: Identifying key criteria using a spatially explicit individual based model

Introduced alien species are the second most important threat to global biodiversity after habitat loss. The American mink Mustela vison has been introduced to several countries and is threatening a number of native species worldwide. We developed a spatially explicit and individual based model as a planning tool to identify key criteria for the implementation of trapping campaigns as a way to control open American mink populations. We first predicted the minimum effort required to reduce populations of mink below a certain threshold and the best time of year in which to trap mink to minimise their numbers. We then employed this methodology to predict the best trapping strategy to ensure the long-term survival of the water vole Arvicola terrestris, one of the species most endangered by the spread of the mink in the UK. We also applied the mink and water vole population models to rationalise a set of observed data in an area of 50 × 30 km in the Upper Thames (UK). The model predicted that it is necessary to remove mink for at least 3 months every year and that a mixed strategy of trapping during the mating, late dispersal and winter seasons is best for keeping mink at low densities. Concentrating trapping during the late dispersal and winter seasons is instead best for ensuring the long-term survival of water voles. Targeting immigrating juvenile mink as well as reproductive adults is important. The model also showed that trapping efficiency might be an important factor to consider when choosing periods in which to trap.     

American mink control on inland rivers in southern England: An experimental test of a model strategy

Invasive species are an important driver of global biodiversity loss. Under international legislation, the UK has an obligation to eradicate or to control the alien, invasive American mink. Using a large-scale field experiment, we tested the effectiveness of a specified mink removal strategy, identified through earlier modelling work, in reducing the relative abundance of mink. We found that mink removal could be effective in reducing mink populations with four months or less of trapping per year, over only 2-3 years, but that for small sites (c. 20 km) a flexible, reactive approach, coupled with continual monitoring for mink presence is necessary. Survival of reintroduced water voles at four sub-sites within our mink removal sites suggest that the reactive mink removal strategy adopted in this study was sufficient for water vole protection. We discuss the use of an adaptive management approach in local mink management, and consider the wider implications of our results for invasive species control on mainlands.     

Numerical and behavioral responses of waterfowl to the invasive American mink: A conservation paradox

A behavioral response is usually the first adaptation of animals to environmental changes such as the introduction of an alien invasive species, and it seems to be the most important factor in reducing the negative impact of new predators on naive prey. This study examined changes in the abundance and behavior of breeding coots (Fulica atra) and great crested grebes (Podiceps cristatus) in response to invasion of the Mazurian Lakeland (NE Poland) by American mink (Neovison vison), first recorded here in 1984. Feral mink densities varied from 1.0 to 14.6 individuals per 10 km of shoreline. Mink were responsible for the losses of 17–58% of experimental clutches (artificial nests containing chicken eggs) that were deployed. Mink destroyed more experimental clutches than other predators; the effects of mink predation were similar in reeds and on the bank. Radio-tracking showed that mink avoided lake banks near human settlements. Mink have markedly affected the breeding distribution of both species and caused a greater than 20-fold decline in coot numbers. For coots and grebes breeding distribution and nesting success were positively associated with built-up areas along lake shores and also with whether the birds were breeding in colonies. Increased brood survival of coots and grebes near human settlements is a paradox in the sense that a man-made habitat change has provided refuge from an alien predator.     

Habitat management of little terns in Japan’s highly developed landscape

The little tern (Sterna albifrons) is a species of shorebird that nests in colonies on sandy beaches and riverbanks with little vegetation cover. In Japan, the natural breeding habitats of the terns have decreased drastically, and the species is listed under the category “vulnerable” in the Red List of the Government of Japan. To reduce the effects of habitat loss on the terns, conservationists are attempting to create and manage artificial colony sites in highly developed landscapes, such as reclaimed lands located in Tokyo Bay, central Japan. We present the factors that contribute to breeding success for this species and the habitat characteristics related to nest-site selection in the artificial colony sites along Tokyo Bay. Our data show that around 3 ha of a building rooftop without vegetation cover could sustain more than 2000 little tern nests. We found a strong positive relationship between colony size and hatching success, observing that a colony size of more than 100 lowered predation rates of eggs. On sites built upon white crushed-concrete, nest densities were high and predation rates of eggs were low. We also found that feeding rates were affected by foraging habitats, of which the best types were the shoreline of sandy beaches and mud flats. The 40 km dispersal range of the breeding terns, inferred from distribution data, should be considered when establishing a network of the multiple colony sites.     

Factors associated with survival of reintroduced riparian brush rabbits in California

The riparian brush rabbit (Sylvilagus bachmani riparius) is an endangered species found in dense, brushy habitat in the California’s Central Valley. We implemented a reintroduction program to bolster populations at a Federal Wildlife Refuge and to assess factors influencing mortality and subsequent survival of released individuals. Between July 2002 and July 2005, we reintroduced 325 captive-bred individuals to unoccupied habitat within their historic range using a soft-release strategy and monitored their subsequent survival with radiotelemetry. Longer time in soft-release pens resulted in increased monthly survival. Rabbits were most susceptible to post-release mortality during the first 4 weeks following reintroduction and both body mass and length of time in the soft-release enclosure influenced this relationship. When we controlled for release mortality during this acclimation period, subsequent monthly survival probabilities were most strongly influenced by release year (year 1 vs. years 2 and 3) and by a catastrophic flooding event; length of time in the soft-release enclosure remained an important variable in longer-term survival. Cause of mortality was unknown for the majority of deaths (61.9%), but predation (including presumptive predation) was the greatest known cause of death in translocated rabbits (26.4%). Reintroduction programs should employ an adaptive management approach with ongoing monitoring of target animals and concurrent analysis to allow managers to adjust methods as conditions dictate.     

Predation determines the outcome of 10 reintroduction attempts in arid South Australia

Ten reintroduction attempts were conducted in and around the Arid Recovery Reserve in northern South Australia between 1998 and 2008. Five locally-extinct mammal species and one reptile species were reintroduced into a fenced Reserve where cats, foxes and rabbits were excluded. Reintroductions of the nationally threatened greater stick-nest rat, burrowing bettong, greater bilby and western barred bandicoot were all considered successful based on short and medium-term success criteria. These criteria included continued survival after 8 years, increased distribution across the large Reserve and, most importantly, recovery after a drought event. The trial reintroductions of the numbat and woma python into the Reserve were unsuccessful due to predation by native avian and reptilian predators respectively. Outside the Reserve, where cats and foxes were present but controlled through poison baiting, reintroduction attempts of the greater bilby and burrowing bettong were unsuccessful. High mortality was attributed to cat and fox predation with dingoes also contributing to post-release mortality in bettongs. However, a reintroduction of burrowing bettongs into a fenced area with low rabbit and cat abundance has, to-date, met short-term and medium-term success criteria. Results suggest that the absence or severe restriction of exotic mammalian predators was the critical factor responsible for the success of the mammal reintroductions. Determining thresholds of predator activity below which successful reintroduction of threatened species can occur, are needed to improve the science of reintroduction biology in Australia.     

Captive-reared burrowing owls show higher site-affinity, survival, and reproductive performance when reintroduced using a soft-release

A remaining challenge for animal reintroductions is how best to aid individuals transitioning from captivity into the wild. We tested two techniques – ‘soft’ vs. ‘hard’ release – in a 7-year study on endangered Western burrowing owls (Athene cunicularia hypugaea) in British Columbia, Canada. Traditionally, captive-reared yearling adults were released into the wild, as pairs, directly from artificial burrows (hard-release). Only 46% of 201 owls released in this way (2001–2004) stayed at release sites. To test for improved success, soft-releases were performed in the same manner as concurrent hard-releases (2005–2007), except above-ground enclosures confined each soft-release pair to the vicinity of their burrow for 2 weeks before complete release. Of 140 soft-released owls, 86% stayed at release sites; whereas, 66% of 100 hard-released owls stayed. Breeding-season survival was 70% for soft-released owls vs. 50% for hard-released owls. On average, soft-released owls produced 50% more fledglings than concurrently hard-released owls. Post-fledging survival was 69% and first-year return rate was 7.0% for fledglings from soft-released parents, compared to 50% and 4.0% for fledglings from hard-release parents. Ultimately, 0.17 offspring were recruited into the local wild breeding population per soft-released pair, compared to 0.05 local recruits per hard-released pair. We recommend that enclosure-based soft-releases be used for reintroduction of burrowing owls in British Columbia and elsewhere in North America. We encourage other tests for improved release success with soft-releases, particularly for species with a high tendency for dispersal or those likely to experience significant predation pressure.