Given current extinction trends, the number of species requiring conservation breeding programs (CBPs) is likely to increase dramatically. To inform CBP policies for threatened terrestrial vertebrates, we evaluated the number and representation of threatened vertebrate species on the IUCN Red List held in the ISIS zoo network and estimated the complexity of their management as metapopulations. Our results show that 695 of the 3,955 (23%) terrestrial vertebrate species in ISIS zoos are threatened. Only two of the 59 taxonomic orders show a higher proportion of threatened species in ISIS zoos than would be expected if species were selected at random. In addition, for most taxa, the management of a zoo metapopulation of more than 250 individuals will require the coordination of a cluster of 11 to 24 ISIS zoos within a radius of 2,000 km. Thus, in the zoo network, the representation of species that may require CBPs is currently low and the spatial distribution of these zoo populations makes management difficult. Although the zoo community may have the will and the logistical potential to contribute to conservation actions, including CBPs, to do so will require greater collaboration between zoos and other institutions, alongside the development of international agreements that facilitate cross-border movement of zoo animals. To maximize the effectiveness of integrated conservation actions that include CBPs, it is fundamental that the non-zoo conservation community acknowledges and integrates the expertise and facilities of zoos where it can be helpful.
Citation: Conde DA, Colchero F, Gusset M, Pearce-Kelly P, Byers O, Flesness N, et al. (2013) Zoos through the Lens of the IUCN Red List: A Global Metapopulation Approach to Support Conservation Breeding Programs. PLoS ONE 8(12): e80311. doi:10.1371/journal.pone.0080311
Editor: Matt Hayward, Bangor University, United Kingdom
Received: June 27, 2013; Accepted: October 1, 2013; Published: December 11, 2013
Copyright: © 2013 Conde et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: These authors have no support or funding to report.
Competing interests: The authors have declared that no competing interests exist.
The conservation status of known biodiversity has undergone a worrying decline in the last few decades , . If the present trends continue, the conservation community will be challenged with a large number of species for which there is no viable conservation outcome. As a result, conservation breeding programs (CBPs) may offer the only feasible option to avoid the extinction of particular species until appropriate habitat can be found or restored –. The need for well-managed CBPs, including those supported by the zoo community, has never been greater, but the space and logistical demands for managing CBPs could exceed the current capacity of the zoo community , . Although we know that 15% of threatened vertebrate species are represented in zoos , we still do not have an overview of the number and distribution of these species within the global zoo network. Therefore, to establish effective responses to species' CBPs, it is important to determine the current representation of threatened species within zoos and have a general understanding of the complexity of managing them within the global zoo network.
The development of CBPs is challenging and has received criticism due to problems including hybridization risk, high costs and the diversion of resources away from habitat protection –. Because of these factors, CBPs have usually been considered only as a last resort and, as a result, are normally only implemented when populations dwindle to fewer than 20 individuals (e.g. whooping crane, Grus americana ). Still, CBPs have played a major role in 13 of the 68 species that have shown a status improvement in IUCN Red List reassessments , . Moreover, for mammals, captive breeding together with reintroduction programs and hunting restriction have been the most effective conservation actions, since ensuring protected areas alone has not been enough . As a result, in most cases effective conservation plans require the integration of a range of management options where CBPs could be necessary. For example, Martin et al.  compared the failure of conservation intervention for the recently extinct Christmas Island pipistrelle (Pipistrellus murrayi) with the potentially successful intervention in the case of the orange-bellied parrot (Neophema chrysogaster), where conservation actions, including CBPs, were implemented in a timely manner and coordinated among several institutions.
As threats to biodiversity increase, the management of at-risk species requires a spectrum of interventions that can range from habitat protection to the establishment of CBPs with the eventual aim of reintroduction into the wild , . Therefore, differentiating between in situ and ex situ management is becoming irrelevant , . Instead, a continuum of management practices exists, which ranges from truly wild and self-sustaining populations to managed populations dependent on a certain level of human care for persistence. For example, the IUCN SSC Conservation Breeding Specialist Group has recently launched an ambitious conservation framework called the One Plan approach, which promotes participation between different institutions and stakeholders with the aim to consider all populations of a particular species, both inside and outside their natural range, under a unified management plan . The zoo community is in an ideal position to support and further contribute to develop such conservation programs because they are involved in both captive breeding and field conservation programs. For example, zoos have extensive knowledge of the husbandry, behavior and veterinary procedures required to develop CBPs , and members of the World Association of Zoos and Aquariums (WAZA) are, collectively, the third largest financial supporter of species conservation in their natural habitats (providing US$350 million/year), while also being actively involved in many of those projects .
To understand the space devoted to threatened species and the potential complexity of managing CBPs as zoo-held metapopulations, we evaluated the representation of the IUCN Red List threatened species held within the zoo network. We used data published from Conde et al. , processed from the International Species Information System (ISIS) organization. ISIS is a network of more than 800 zoos and aquariums that shares information about ∼2.5 million individuals among the member institutions. This information system makes it possible to analyze the amount of space devoted to threatened species and its spatial distribution. Therefore, ISIS is a key institution to assess the potential of zoos to develop CBPs. To understand if the number of threatened species in zoos is the result of a sound prioritization or if it has been more an opportunistic process, we analyzed whether the representation of threatened species per taxonomic order is significantly higher to what would be expected if species were randomly collected (i.e. without collection planning). In addition, to appraise the approximate complexity of managing zoo-held species as metapopulations, we estimated the distance between zoos clusters at which the probability of reaching an average metapopulation size of 50, 100 and 250 individuals is maximized. With these analyses we can better infer how complex it is to reach increasing population sizes between ISIS zoos, assuming that CBPs could be more successful when population sizes are larger and clustered in zoos at closer proximity. Since the clustering of zoos could facilitate special treaties to move individuals across borders. Based on our results, we discuss the potential of zoos to contribute towards CBPs for terrestrial vertebrates and make some policy recommendations.
Representation of threatened species by order in the ISIS network
To determine whether the number of threatened species in ISIS zoos is different from what would result from a random sample, we carried out the following analyses. For every taxonomic order i of terrestrial vertebrates there is a given number ni of species worldwide, out of which a fraction mi is known to be threatened. Similarly, we know the number of species for every order zi that are represented in ISIS zoos and the subset wi of those that are threatened. We developed a Monte Carlo algorithm to understand which orders have a number of threatened species in ISIS zoos significantly different from a random sample of species in the wild. This algorithm was based on an iterative procedure that, at every step s, randomly sampled, without replacement, zi species within every order from the worldwide list. From these randomly sampled species, the algorithm counted the number of species classified as threatened. The algorithm then calculated the indicator yi,j,s that assigned 1 if a number j of threatened species were sampled and 0 otherwise, such that j = 0, …, min(zi, mi). The probability of randomly finding j threatened species is calculated aswhere S is the total number of iterations. Since the algorithm ran for several thousand iterations, we were able to construct a distribution (i.e. empirical probability mass function, PMF) of the number of threatened species per order that could have been found in ISIS zoos if they had sampled species at random from the wild. The probability in the PMF that corresponds to the real number of threatened species in zoos () was used as an analogue to the P-value. Those orders that had a number of threatened species that matched the lower bounds of the PMF (i.e. P-value of 0.05) were classified as being significantly under-represented. Similarly, those that matched upper bounds of the PMF (i.e. above the 0.95 quantile) were catalogued as being over-represented. The remaining species could not be distinguished from a random sample.
Managing metapopulations of threatened species in the ISIS network
To understand the complexity of maintaining CBPs of threatened species across the ISIS zoo network, we developed a second algorithm to find the optimal radial distance from any given zoo at which the probability of finding a metapopulation size of at least 50, 100 or 250 recruited living individuals was highest relative to that radial distance. The algorithm was based on a Monte Carlo procedure that, at each step s, a zoo j was selected at random, then found all zoos zj,r located at a distance r from zoo j, as well as the number of zoos within that radial distance that held species i. We labeled the zoos included within that radius as a cluster, Kj,r. The algorithm then counted how many individuals of a given threatened species i were included in the cluster, such thatwhere ni,k is the population size for species i in zoo k. At each cluster, we assigned the indicatorwhere m is a pre-established metapopulation threshold (i.e. 50, 100 or 250 individuals). We chose 50, 100 and 250 individual metapopulation thresholds because a population size of 50 has been historically considered as a minimum viable population , and 250 individuals is the threshold defined by the IUCN Red List as a “very small wild population” in the Critically Endangered category . Moreover, very few threatened species in ISIS zoos have populations for which a higher threshold is possible. For instance, only 44 species (6.3%) have more than 1,000 individuals (see below). We considered 100 individuals as an arbitrary intermediate between these two values.
This procedure was repeated for 2,000 iterations and then the algorithm was repeated, increasing the radial distance by 100 km until a maximum radius of 10,000 km was reached. From the 2,000 iterations for each radius, the algorithm calculated the probability of finding at least m individuals for species i and radius r aswhere S is the total number of steps (2,000) in the Monte Carlo procedure. Here, we replaced the subscript j with s to indicate that zoos were chosen at random with replacement. Similarly, we calculated the average number of zoos for each species at each distance interval as
We repeated this procedure for radii ranging from 0 to 10,000 km, using 100 km increments. We excluded species for which the minimum population size was never reached, even when including all ISIS zoos.
For each class C and metapopulation threshold m, we calculated the average ratio between probability pi,r,m and distance dr for radius r aswhere IC is the total number of threatened species in class C that was included in the analysis. Finally, we found the optimal cluster, this is the optimal radius and the optimal number of zoos , where the ratio RC,r,m was highest for each class and population threshold m. These optimal clusters imply that, relative to that radius, the probability of reaching a metapopulation of at least m individuals is highest. For simplicity, we present the results pooled by taxonomic class.
Collectively, in 2011, the 837 ISIS zoos held 3,955 species of non-domestic terrestrial vertebrates (Table 1). Of these, more than half (58%) were birds, one quarter (25%) were mammals, 11% were reptiles and 6% were amphibians (Figure 1). Twenty-three percent (691/3,955) of the species in ISIS zoos belong to a threatened category (Table 2). Within each class, the percentage of threatened species varies widely among each order, ranging between 8% for birds (195/2,308) and 27% for mammals (262/978). ISIS zoos held a total of 455,317 individuals of non-domestic terrestrial vertebrates, of which 22% (91,063/455,317) belong to a threatened species (Table S1).
The upper panel summarizes the representation of species for each taxonomic class of terrestrial vertebrates while the four lower panels summarize representation at the taxonomic order level within each class. The length of each bar is proportional to the number of species held, and each bar is color-coded by class. The darker shaded region of each bar represents the number of species that fall into the three IUCN Red List threatened categories (Vulnerable, Endangered and Critically Endangered). The number of species in each category is indicated to the right of each bar, and the percentage of these that are threatened is given in parentheses.
Representation of threatened species by order in the ISIS network
Our analysis of the number of threatened species in ISIS zoos, broken down by order, shows that most collections are not distinguishable from what would be expected if the species were selected at random (Figure 2, Tables S2, S3, S4, S5). Exceptions occur in mammals in the order Dasyuromorphia (Australian carnivorous marsupials) and in reptiles for Testudines (turtles): ISIS zoos hold 50% and 79%, respectively, of the order's threatened species. On the other hand, threatened species were under-represented in zoos for the mammalian orders Eulipotyphyla (insectivores) and Rodentia (rodents). For birds, threatened species were under-represented in nine of the 25 orders held in zoos, whereas for amphibians this was only the case for Caudata (salamanders).
Each point represents an order that was significantly different from the expected value if zoo collections were taken at random. The size of the point corresponds to the number of threatened species for each order. Labels correspond to: a) Caudata; b) Testudines; c) Anseriformes; d) Charadriiformes; e) Columbiformes; f) Coraciiformes; g) Falconiformes; h) Passeriformes; i) Pelecaniformes; j) Psittaciformes; k) Strigiformes; l) Dasyuromorphia; m) Eulipotyphyla; n) Rodentia.
Managing metapopulations of threatened species in the ISIS network
For the threatened amphibian species in ISIS zoos, 27% reach a metapopulation size threshold of >50 individuals. Almost half of the threatened species of mammals, birds and reptiles in ISIS zoos reach the same threshold (44%, 47% and 43%, respectively), and 18% of the threatened mammals reach a threshold above 250 individuals (Table 3). However, many of those species are distributed among numerous zoos, thus if those species are not managed as a metapopulation, their conservation potential will be greatly reduced.
As expected, the optimal distance radii and the number of zoos required to maximize the probability of reaching a given population increase with the metapopulation size threshold (>50, >100 and >250 individuals) (Figure 3). However, the magnitude of the increase varies among classes; for birds and reptiles, the difference between optimal distances is just 200 km and 300 km, respectively, while for mammals and amphibians, the differences are up to 900 km (Table 4).
When the ratio is highest, we obtain the optimal radial distance between zoos and the optimal probability of finding a metapopulation size above the threshold. This is, at that ratio the probability of finding the metapopulation is highest with respect to the zoo cluster radial distance. For display purposes all ratios were multiplied by 1,000.
Furthermore, at the optimal radial distances, the probability that a cluster of zoos has more than 50, 100 or 250 individuals of a given threatened species ranges from 0.16 to 0.42 (Table 4). However, only 3.5% to 10.2% of zoos within an optimized cluster have individuals of the species. Moreover, these percentages vary considerably among the four classes of terrestrial vertebrates.
Representation of threatened species by order in the ISIS network
Birds and mammals make up the largest proportion of terrestrial vertebrates held in zoos. Although the majority of zoo collection plans were not originally focused on holding threatened species , 23% of their collections are currently devoted to them. However, for most of the taxonomic orders, our results show that representation of threatened species is not different from what would be expected if species were selected at random. Broken down by class, it is clear that threatened birds make up the lowest proportion in the zoo network. This may be explained by factors including: i) some species are difficult to breed in captivity (e.g. seabirds), ii) some species have specialized dietary requirements (e.g. insectivores), and iii) import/export restrictions, such as those added in the wake of the SARS (severe acute respiratory syndrome) epidemic, which make it challenging to manage birds across borders and to import new species.
Mammalian and reptilian zoo collections include the highest proportion of threatened species; however, only the orders Dasyromorphia and Testudines are significantly over-represented. The Dasyromorphia are a particular focus of CBP efforts in Australian zoos . However, despite the interest in this group, six threatened Dasyromorphia species are not yet represented in zoos. The high representation of threatened Testudines is partly because many zoos serve as rescue centers for confiscated individuals. The Turtle Survival Alliance has played a key role in promoting the conservation of turtles and has been working on linking zoos and governmental institutions to ensure the rescue of animals from the illegal trade . However, since many of those individuals come from the illegal trade, it is hard to include them as part of CBPs because, in many cases, their origins are uncertain .
For amphibians, the relatively low proportion of threatened Caudata in zoos may reflect their cryptic behavior and small size , which makes them difficult to display. In general, the small number of amphibian species may also be due to practical issues such as the difficulty in obtaining permits to transport individuals. Although amphibian collections in zoos and zoo-supported centers have significantly increased in the last 10 years , , zoos still only hold 3% of the world's threatened amphibians . This emphasizes the need for zoos to increase their contribution towards amphibian CBPs either as part of their collections or by further contributing to the development of breeding centers in their local areas . With this in mind, the Amphibian Ark emerged with a strong zoo component with the mission of ensuring the global survival of amphibians, focusing on those that cannot currently be safeguarded in nature and where zoos can play a key role .
Managing metapopulations of threatened species in the ISIS network
Zoos hold ∼15% of the world's threatened terrestrial vertebrates . However, in their collections, 23% of ISIS zoos species (691/3,955) belong to a threatened category. Nevertheless, most of their populations are small and are distributed across the zoo network. Therefore, for the zoo community, one of the main challenges of managing their threatened species in CBPs is the complexity of moving individuals across borders and the coordination of conservation efforts among zoos and other institutions at a global level. The enormity of this task is clear from our results. For example, the optimal radius for finding a metapopulation size of more than 250 individuals for a given threatened species is 1,700 km for reptiles and 2,200 km for mammals. Within these optimized clusters the number of zoos that hold a given threatened species is low compared with the number of zoos available within the clusters. For example, on average only 13% (17/227) of all the zoos within a cluster hold a given threatened reptile species. This implies that, under current conditions, to manage a metapopulation above 250 individuals requires an optimized cluster of ∼20 zoos that could be up to 4,000 km apart. Given this complexity, it is not surprising that most zoo populations of threatened species are not managed as metapopulations , nor that most are not yet sustainable in the long term .
We found that, on average, for threatened species in the ISIS network, fewer than 10% of zoos within optimized clusters hold a particular species. Therefore, it would be possible to improve the network within an optimized cluster by increasing the number of zoos that contain individuals of a focal species, managing these collections as a single metapopulation and potentially reducing the distances between zoos. This level of organization could result in zoos focusing on particular CBPs for fewer taxa, rather than having a small number of individuals of many threatened species. This is particularly important since specialization has been shown to increase breeding success . This observation does not mean that zoos should shift their entire collections towards one or a few at-risk species, since responsible zoos have other conservation goals such being centers for education and research . Rather, it means that zoos within a particular region can most efficiently increase their conservation contribution by developing collectively managed CBPs devoted towards a smaller number of focal species.
The proportion of threatened species that exceed a threshold metapopulation size of 250 individuals is rather low, ranging from only 9% for birds to 18% for mammals. However, the percentages of species reaching the threshold of >50 individuals range from 27% for amphibians to ∼45% for the other three classes. ISIS zoos have only a small number of threatened species for which population sizes are above 1,000 individuals. Although this number has been suggested as an appropriate threshold over which genetic diversity should be maintained , most of the species that have been recovered from CBPs come from populations below 30 to 20 individuals (see ). In this sense, it is important not to underestimate the potential of some of these ISIS populations. Nonetheless, the zoo community should aim at providing populations to CBPs that can ensure genetic and demographic sustainability . Additionally, CBPs should not be implemented only when species have reached dramatically low numbers, at which point their chances of success are lower . Our results stress that, for many species, appropriate management and coordination within an optimized cluster of zoos can potentially increase their numbers to at least 250 individuals. In addition, it is expected that these population sizes will be larger if we include non-ISIS member institutions. However, it is important to stress that the successful management of CBPs as metapopulations requires the collaboration and coordination of zoos within a global network such as ISIS.
Management and policy recommendations
- The zoo community should identify potential zoo clusters for the conservation of prioritized species. Particularly for small-bodied species, zoos could potentially hold a large number of individuals within a particular region. Although the clusters should ideally be in close proximity to the species' natural habitat, most ISIS zoos are currently located away from major biodiversity hotspots  and therefore, zoos' support of breeding centers within the native range may be an option.
- Cluster-level integrated management plans should be implemented to ensure the coordination of CBPs with habitat conservation and other in situ efforts (One Plan approach , . For this, the development of greater coordination among zoos using networks such as ISIS, together with conservation NGOs, academic and governmental institutions, will be essential (see ).
- Species-specific clusters should ideally be replicated to minimize the potential impact of catastrophic events.
- Cross-border management policies for zoos should be modified. In this sense, the management of CBPs by clusters can facilitate treaties for the management of particular focal species. To ensure sustainable metapopulations it will require more than the coordination and will of the zoo and conservation community. Although we acknowledge the need for public health management and vigilance against illegal trade, the development of cross-border management policies will be key to achieving successful CBPs. For example, CITES could award special permits to facilitate movement of targeted individuals within CBP clusters.
Global change represents an unprecedented challenge for the maintenance of biodiversity –. It is expected that even under the most optimistic impact and adaptation scenarios, a great number of species may require the integration of a suite of conservation actions, including CBPs. Furthermore, species that have no likelihood of in situ persistence for the foreseeable future represent an additional conservation challenge. For example, under current global warming trends, most polar and some montane species are likely to fall into this category , in addition to species whose habitat will be lost by urbanization . As a result, deciding which species could be part of successful CBPs and which institutions should modify their collections to become part of a particular cluster needs to be the result of a sound prioritization approach. Furthermore, simply holding these species, even in numbers above the minimum 50 individuals, is not sufficient , and CBPs need to be integrated with other aspects of species conservation, such as habitat protection and restoration, eradication of invasive species and population management –, , . Our results show that, in the zoo network, the representation of species that may require CBPs is currently low for most taxa and the spatial distribution of these zoo populations makes management difficult. However, the zoo network already devotes 23% of its collections to threatened species; for mammals, 18% of those reach population sizes above 250 individuals. Zoos in collaboration with other institutions have already saved a number of species from extinction, but it has been mostly opportunistic rather than strategic. If zoos collectively focus on their strength as a global network, they have the potential for the development of integrated conservation programs that include CBPs. To maximize effectiveness, the collaboration of the global zoo network with governmental institutions, regional and international trade authorities, NGOs and academia should be fostered. Such collaborations are already underway, and termed a One Plan approach . However, it is essential to strengthen these institutions' contributions, include special international treaties and collaborations to help slow down current extinction trends.
Number of individual animals in ISIS zoos in each of the IUCN Red List categories for each taxonomic class of terrestrial vertebrates.
Representation of threatened mammal species for each taxonomic order in ISIS zoos.
Representation of threatened bird species for each taxonomic order in ISIS zoos.
Representation of threatened reptile species for each taxonomic order in ISIS zoos.
Representation of threatened amphibian species for each taxonomic order in ISIS zoos.
Thanks to the IUCN SSC Conservation Breeding Specialist Group for facilitating the discussions that set the direction of this paper; to ISIS member institutions for making their basic data available in the ISIS portal; to Alexander Scheuerlein and Zjef Pereboom for useful comments on the manuscript; and to Robert Wiese, William van Lint, Jenny Gray and Jeffrey Bonner for sharing their perspectives and expertise on zoo collections. Additionally, we would like to thank the editor and two external reviewers for helpful suggestions and detailed comments that significantly improved our manuscript.
Conceived and designed the experiments: DAC FC ORJ. Analyzed the data: DAC FC. Wrote the paper: DAC FC PPK MG OB NF RKB ORJ.
- 1. Collen B, Loh J, Whitmee S, McRae L, Amin R, et al. (2009) Monitoring change in vertebrate abundance: the Living Planet Index. Conservation Biology 23: 317–327. doi: 10.1111/j.1523-1739.2008.01117.x
- 2. Butchart SHM, Walpole M, Collen B, van Strien A, Scharlemann JPW, et al. (2010) Global biodiversity: indicators of recent declines. Science 328: 1164–1168. doi: 10.1126/science.1187512
- 3. Conde DA, Flesness N, Colchero F, Jones OR, Scheuerlein A (2011) An emerging role of zoos to conserve biodiversity. Science 331: 1390–1391. doi: 10.1126/science.1200674
- 4. Pritchard DJ, Fa JE, Oldfield S, Harrop SR (2011) Bring the captive closer to the wild: redefining the role of ex situ conservation. Oryx 46: 18–23. doi: 10.1017/s0030605310001766
- 5. Conway WG (2010) Buying time for wild animals with zoos. Zoo Biology 30: 1–8. doi: 10.1002/zoo.20352
- 6. Bowkett AE (2009) Recent captive-breeding proposals and the return of the ark concept to global species conservation. Conservation Biology 23: 773–776. doi: 10.1111/j.1523-1739.2008.01157.x
- 7. Balmford A, Leader-Williams N, Green MJB (1995) Parks or arks: where to conserve threatened mammals? Biodiversity and Conservation 4: 595–607. doi: 10.1007/bf00222516
- 8. Snyder NFR, Derrickson SR, Beissinger SR, Wiley JW, Smith TB, et al. (1996) Limitations of captive breeding in endangered species recovery. Conservation Biology 10: 338–348. doi: 10.1046/j.1523-1739.1996.10020338.x
- 9. Rahbek C (1993) Captive breeding – a useful tool in the preservation of biodiversity? Biodiversity and Conservation 2: 426–437. doi: 10.1007/bf00114044
- 10. Parr MJ, Bennun L, Boucher T, Brooks T, Aucca Chutas C, et al. (2009) Why we should aim for zero extinction. Trends in Ecology and Evolution 24: 181. doi: 10.1016/j.tree.2009.01.001
- 11. Hoffmann M, Hilton-Taylor C, Angulo A, Böhm M, Brooks TM, et al. (2010) The impact of conservation on the status of the world's vertebrates. Science 330: 1503–1509.
- 12. Conde DA, Flesness N, Colchero F, Jones OR, Scheuerlein A (2011) Zoos and captive breeding – response. Science 332: 1150–1151. doi: 10.1126/science.332.6034.1150
- 13. Hayward MW (2011) Using the IUCN Red List to determine effective conservation strategies. Biodiversity and Conservation 20: 2563–2573. doi: 10.1007/s10531-011-0091-3
- 14. Martin TG, Nally S, Burbidge AA, Arnall S, Garnett ST, et al. (2012) Acting fast helps avoid extinction. Conservation Letters 5: 274–280. doi: 10.1111/j.1755-263x.2012.00239.x
- 15. Redford KH, Jensen DB, Breheny JJ (2012) Integrating the captive and the wild. Science 338: 1157–1158. doi: 10.1126/science.1228899
- 16. Byers O, Lees C, Wilcken J, Schwitzer C (2013) The One Plan approach: the philosophy and implementation of CBSG's approach to integrated species conservation planning. WAZA Magazine 14: 2–5.
- 17. Ryder OA, Feistner ATC (1995) Research in zoos: a growth area in conservation. Biodiversity and Conservation 4: 671–677. doi: 10.1007/bf00222522
- 18. Gusset M, Dick G (2010) “Building a Future for Wildlife?” Evaluating the contribution of the world zoo and aquarium community to in situ conservation. International Zoo Yearbook 44: 183–191. doi: 10.1111/j.1748-1090.2009.00101.x
- 19. Shaffer ML (1981) Minimum population sizes for species conservation. BioScience 31: 131–134. doi: 10.2307/1308256
- 20. IUCN Red List of Threatened Species (2013) Available: http://www.iucnredlist.org. Accessed 1 April 2013.
- 21. Fa JE, Funk SM, O'Connell D (2011) Zoo Conservation Biology. Cambridge: Cambridge University Press.
- 22. Harley D (2012) The application of Zoos Victoria's “Fighting Extinction” commitment to the conservation of Leadbeater's possum Gymnobelideus leadbeateri. The Victorian Naturalist 129: 175–180.
- 23. Turtle Survival Alliance (2013) Available: http://www.turtlesurvival.org. Accessed 1 June 2013.
- 24. Duellman WE, Trueb L (1994) Biology of Amphibians. Baltimore, MD: Johns Hopkins University Press.
- 25. Browne RK, Wolfram K, García G, Bagaturov MF, Pereboom ZJJM (2011) Zoo-based amphibian research and conservation breeding programs. Amphibian and Reptile Conservation 5: 1–14.
- 26. Zippel K, Johnson K, Gagliardo R, Gibson R, McFadden M, et al. (2011) The Amphibian Ark: a global community for ex situ conservation of amphibians. Herpetological Conservation and Biology 6: 340–352.
- 27. Amphibian Ark (2013) Available: http://www.amphibianark.org. Accessed 1 June 2013.
- 28. Traylor-Holzer K (2012) Identifying gaps and opportunities for inter-regional ex situ species management. WAZA Magazine 12: 30–33.
- 29. Lees CM, Wilcken J (2009) Sustaining the ark: the challenges faced by zoos in maintaining viable populations. International Zoo Yearbook 43: 6–18. doi: 10.1111/j.1748-1090.2008.00066.x
- 30. Traill LW, Bradshaw CJ, Brook BW (2007) Minimum viable population size: a meta-analysis of 30 years of published estimates. Biological Conservation 139: 159–166. doi: 10.1016/j.biocon.2007.06.011
- 31. Lacy RC (2013) Achieving true sustainability of zoo populations. Zoo Biology 32: 19–26. doi: 10.1002/zoo.21029
- 32. Brook BW, Sodhi NS, Bradshaw CJA (2008) Synergies among extinction drivers under global change. Trends in Ecology and Evolution 23: 453–460. doi: 10.1016/j.tree.2008.03.011
- 33. Thomas CD, Cameron A, Green RE, Bakkenes M, Beaumont LJ, et al. (2004) Extinction risk from climate change. Nature 427: 145–148. doi: 10.1038/nature02121
- 34. Bellard C, Bertelsmeier C, Leadley P, Thuiller W, Courchamp F (2012) Impacts of climate change on the future of biodiversity. Ecology Letters 15: 365–377. doi: 10.1111/j.1461-0248.2011.01736.x
- 35. Shrestha AB, Devkota LP (2010) Climate Change in the Eastern Himalayas: Observed Trends and Model Projections. Kathmandu: International Centre for Integrated Mountain Development.
- 36. Seto KC, Güneralp B, Hutyra LR (2012) Global forecasts of urban expansion to 2030 and direct impacts on biodiversity and carbon pools. Proceedings of the National Academy of Sciences of the USA 109: 16083–16088. doi: 10.1073/pnas.1211658109