Taxonomic bias in reintroduction projects

Transcription

1 Animal Conservation (2005) 8, C 2005 The Zoological Society of London. Printed in the United Kingdom DOI: /S Taxonomic bias in reintroduction projects Philip J. Seddon 1,3,, Pritpal S. Soorae 2,3 and Frédéric Launay 2,3 1 Department of Zoology, University of Otago, P.O. Box 56, Dunedin, New Zealand 2 Environmental Research and Wildlife Development Agency, P.O. Box 45553, Abu Dhabi, United Arab Emirates 3 World Conservation Union (IUCN) Reintroduction Specialist Group; web address: (Received 2 February 2004; resubmitted 17 June 2004; accepted 27 June 2004) Abstract Taxonomic bias has been documented in general science and conservation research publications. We examined whether taxonomic bias is similarly severe in actual conservation programmes as indicated by the focus of species reintroduction projects worldwide. We compiled a database of reintroduction projects worldwide, yielding a total of 699 species of plants and animals that are the focus of recent, current or planned reintroductions. Using IUCN (World Conservation Union) data for total numbers of known species worldwide, we found that vertebrate projects were over-represented with respect to their prevalence in nature. Within vertebrates, mammals and, to a lesser extent, birds, were over-represented, whereas fish were underrepresented. This over-representation extended to two mammal orders, artiodactylids and carnivores, and to four bird orders, anseriforms, falconiforms, gruiforms and galliforms. For neither mammals nor birds was reintroduction project bias related to any differences between orders in vulnerability to threat. Bird species that are the focus of reintroduction efforts are more likely to be categorised as Threatened than expected on the basis of the distribution of all known species over all threat categories, however, nearly half of all bird species being reintroduced are classified as Least Concern. The selection of candidates for reintroduction programmes is likely to consider national priorities, availability of funding and local community support, over global conservation status, While a focus on charismatic species may serve to garner public support for conservation efforts, it may also divert scarce conservation resources away from taxa more in need of attention. INTRODUCTION Whole-organism research in taxonomy and ecology has been shown not to be proportional to the frequency of organisms in nature (Gaston & May, 1992; Bonnet, Shine & Lourdais, 2002). Recent reviews examined whether taxonomic bias was less evident in conservation research, since conservationists advocate an integrated approach to biodiversity preservation (Clark & May, 2002a,b). Examination of papers published during in two of the leading conservation research journals, Conservation Biology (USA) and Biological Conservation (UK), revealed that taxonomic bias pervades the conservation literature also (Clark & May, 2002a). Publication of research papers in leading journals is an important indicator of taxonomic bias in conservation focus, but could in turn be biased by the difficulties All correspondence to: Philip Seddon. Department of Zoology, University of Otago, P.O. Box 56, Dunedin, New Zealand. philip.seddon@stonebow.otago.ac.nz of conducting rigorous research on rare or critically endangered species, or by the uneven development of research methodology across faunal groups (Pawar, 2003). Of concern for biodiversity preservation is not only what studies are published, but also what is actively conserved. We wondered if taxonomic bias was less evident in conservation projects. One challenge in categorising conservation activities is that they may encompass wide-ranging approaches with potential benefits for a diversity of species. We considered a reasonable measure of taxonomic focus would be the evaluation of the global taxonomic distribution of species reintroduction programmes. Reintroduction is defined by the World Conservation Union (IUCN) as: An attempt to establish a species in an area which was once part of its historical range, but from which it has been extirpated or become extinct (IUCN, 1998). Reintroduction projects are intensive, expensive, species-focussed conservation actions intended to contribute to the restoration of biodiversity. In this paper we aim to take the previous work showing taxonomic bias in taxonomic, ecological and conservation

2 52 P. J. SEDDON ET AL. research as a basis for asking whether a similar bias pervades species reintroductions. METHODS To evaluate taxonomic bias in reintroduction projects we compiled a catalogue of species that are the focus of recent, current or planned reintroductions. We updated and expanded IUCN Reintroduction Specialist Group (RSG) records by searching available databases and via specific queries to the 300+ RSG members coordinated by the RSG s six taxonomic and seven regional chairs. We compiled a list of 699 species that are the focus of reintroduction efforts. We based our comparisons of the distribution of reintroduction projects with the proportions of known species globally on the same data set used by Clark & May (2002a), the IUCN data for total numbers of described species worldwide (IUCN, 2003). To examine the relative degree of bias among taxonomic groups we calculated the numbers of reintroduction projects per taxon that would be expected if projects were in proportion to known species. We then examined the pattern of the residuals, the difference between the observed and expected values. Because absolute residuals are not comparable when the frequencies vary, we calculated standardised residuals as: (n ij f ij )/ f ij where (n ij f ij ) is the difference between the observed and expected values, standardised by dividing by the square root of the expected value. These standardised residuals, also called Pearson residuals, are directly comparable irrespective of absolute frequencies (Quinn & Keough, 2002). The significance of deviation from expected was further indicated using Chi-squared analysis for comparisons of major groups of organisms. Examination of the distributions at the level of order and below yielded too many zero values for formal contingency table significance testing. To assess whether project bias was a reflection of any inherent bias in the vulnerability of a given taxonomic group we examined the numbers of species listed in the IUCN Red List of Threatened Species within the five top threat categories (IUCN, 2003) as a proportion of numbers of species described for a given taxonomic unit. We then ranked groups according to their deviation from the mean proportion assessed as threatened. Because it is possible that the process of species threat assessment is itself subject to taxonomic bias, we assessed relative vulnerability only for species of mammals and birds, for which assessments are 99% and 100% complete, respectively (IUCN, 2003). Potentially, taxonomic bias in threat assessment could influence reintroduction project bias under the assumption that a species must be listed as threatened before being chosen for reintroduction. We tested this assumption for birds, all species of which have been assessed, by comparing the distribution of reintroduced species with the distribution of all species within each threat category. We used the latest assessment of threat conducted by BirdLife (2004) to assign each reintroduced bird species to a threat category. Taxonomic conventions follow those of the IUCN (2003) unless otherwise stated. RESULTS AND DISCUSSION Taxonomic bias is clearly evident in reintroduction projects, with the distribution of projects between plants, vertebrates and invertebrates being significantly Table 1. Taxonomic bias in reintroduction projects among plants, vertebrates and invertebrates and within vertebrates Number of species % species Number of projects Number of projects Absolute Pearson described described (observed) (expected) residual residual Plants Vertebrates Invertebrates Total Vertebrates Mammals Birds Reptiles Amphibians Fishes Total Figures of number of species described taken from IUCN (2003). Number of projects observed derived from the database of projects compiled by the IUCN Reintroduction Specialist Group. Number of projects expected calculated from the taxonomic distribution of all described species. Absolute residuals are the differences between observed and expected; the Pearson residual is standardised to allow direct comparison between taxa irrespective of absolute frequencies (see the text for details). Positive residuals indicate groups that are over-represented within reintroduction projects, whilst negative residuals indicate under-representation. Values in bold type indicate a moderate to severe degree of bias.

3 Taxonomic bias in reintroduction projects 53 (a) Percentage of known species worldwide (b) Percentage of reintroduction projects Plants 19% Vertebrates 4% Invertebrates 9% Plants 30% Invertebrates 77% Vertebrates 61% Plants Vertebrates Invertebrates Plants Vertebrates Invertebrates Fig. 1. Proportion of major taxa in nature (a) versus the focus of reintroduction projects (b). Table 2. Taxonomic bias in reintroduction projects within Orders of the class Mammalia Number of species % of species Number of projects Number of projects Absolute Pearson described described (observed) (expected) residual residual Artiodactyla Carnivora Cetacea Chiroptera Dasyuromorpha Dermoptera Didelphimorphia Diprotodontia Hyracoidea Insectivora Lagomorpha Macroscelidea Microbiotheria Monotremata Notoryctemorphia Paucituberculata Peramelemorphia Perissodactyla Pholidota Primates Proboscidea Rodentia Scandentia Sirenia Tubulidentata Xenarthra Total Column headings as for Table 1. Positive residuals indicate groups that are over-represented within reintroduction projects, whilst negative residuals indicate underrepresentation. Values in bold type indicate a moderate to severe degree of bias., indicates Pearson residual not calculable since expected value zero.

4 54 P. J. SEDDON ET AL. Table 3. Distribution of reintroduction projects within Families of the Mammal Orders Carnivora and Artiodactyla Number of species Number of projects Number of projects Absolute Pearson described (observed) (expected) residual residual Carnivora Felidae Canidae Herpestidae Hyaenidae Mustelidae Odobnidae Otariidae Phocidae Procyonidae Ursidae Viveridae Total Artiodactyla Suidae Tayassuidae Hippotamidae Camelidae Tragulidae Giraffidae Moschidae Cervidae Antilocapridae Bovidae Total Column headings as for Table 1. Positive residuals indicate groups that are over-represented within reintroduction projects, while negative residuals indicate under-representation. Values in bold type indicate a moderate to severe degree of bias. different from that expected on the basis of the numbers of described species (χ 2 = , d.f. = 2, P < 0.001: Table 1). In comparison with conservation research (Clark & May, 2002a,b) where plant research generally reflects the relative prevalence of plants in nature, plant reintroduction projects were slightly over-represented (30% of projects versus 19% of species: Fig. 1). However, as for research, invertebrate projects were highly underrepresented (9% of projects versus 77% of species), while vertebrate projects were highly over-represented (61% of project versus only 4% of species: Fig. 1). Bias was evident among vertebrates (χ 2 = 269.9, d.f. = 4, P < 0.001) with mammals (41% of projects versus 8% of species) and birds (33% of projects versus 18% of species) being overrepresented (Table 1). Reptile and amphibian projects were generally in proportion to the prevalence of these taxa in nature (17% and 5% of projects versus 14% and 10% of species, for reptiles and amphibians, respectively), whereas fish were under-represented (4% of projects versus 50% of species: Table 1). Closer examination of over-represented taxa revealed that the distribution of reintroduction projects generally followed the prevalence of species within the 26 orders of mammals, with the exception of Artiodactyla and Carnivora, both of which were over-represented as indicated by Pearson residuals of > + 10 (Table 2). These two groups are noteworthy for containing large charismatic species such as Arabian oryx (Oryx leucoryx) and lion (Panthera leo), respectively. A more detailed look at the families of these two orders indicated only slight over-representation of felid species and generally projects reflected prevalence of species in nature (Table 3), thus while there is a bias towards the reintroduction of artiodactylids and carnivores, this is not being driven by any particular group(s) of species. Nor is this bias due to any inherent greater likelihood that these two orders will contain proportionately more threatened species (Table 4). Within the 27 orders of birds there was slight overrepresentation of reintroduction projects on species within the Anseriformes, Falconiformes, Gruiformes and Galliformes (Table 5), although species in these orders are not markedly more likely to be listed as threatened (Table 6). It is possible that this slight bias may arise from a disproportionate interest in species of raptorial and game birds, the latter hinting at sustainable utilisation as an underlying motivation for the restoration of wild populations. Bird species that are the focus of

5 Taxonomic bias in reintroduction projects 55 Table 4. Relative vulnerability to threat among orders of the class Mammalia Deviation from Number threatened Total assessed % threatened the mean Rank Artiodactyla Carnivora = Cetacea = Chiroptera = Dasyuromorpha = Dermoptera = Didelphimorphia Diprotodontia = Hyracoidea Insectivora Lagomorpha Macroscelidea = Microbiotheria = Monotremata = Notoryctemorphia = Paucituberculata Peramelemorphia = Perissodactyla Pholidota = Primates Proboscidea = Rodentia Scandentia = Sirenia = Tubulidentata = Xenarthra Total mean = 43 Number threatened is the number of species in each order classified within the IUCN Red List (IUCN, 2003) as Extinct, Critically Endangered, Endangered, or Vulnerable. The total assessed excludes species listed as Data Deficient within the IUCN list. The deviation from the mean is the difference between the proportion of a given order listed as threatened and the overall mean proportion threatened derived from all mammalian orders. Positive deviations indicate that a given order contains relatively more threatened species. Values in bold type indicate extreme positive deviations. reintroduction efforts are more likely to be listed as threatened than any random selection (χ 2 = 157, d.f. = 5, P < 0.001), however, listing as Extinct in the Wild (EW), Critically Endangered (CR), Endangered (En) or Vulnerable (Vu) is not a prerequisite to reintroduction, since nearly 50% of bird species that are being reintroduced are listed as Least Concern (Table 7). It seems likely that the selection of reintroduction candidates is driven more by national priorities, the availability of funding and the level of support from non-governmental organizations (NGOs) and local communities, than by considerations of global conservation status. Bias in the focus of intensive species-centred conservation efforts has even greater significance in the light of the disparity in extinction risk across taxonomic groups. Although globally only birds and mammals have been subject to a comprehensive assessment of conservation status, analyses of organisms in the USA found that birds and mammals had the lowest extinction risk levels among over species examined (Stein, Master & Morse, 2002). It is likely that the figures from the USA will be consistent with global patterns. This sustained and pervasive focus on mammals and, to a lesser extent, on birds, appears to arise from the bias in funding and research activities of the majority of conservation organisations (Czech, Krausman & Borkhataria, 1998). Such bias in conservation action is at odds with the increasing emphasis being placed on the need to halt declines in biodiversity through active ecosystem-level restorations. The obvious question these findings beg is: how critical is it that we reduce taxonomic bias in reintroductions? After all, it is unrealistic that we should suddenly shift our focus to the reintroduction of under-represented taxa such as spiders and beetles. Species reintroductions that focus on charismatic vertebrates could serve to garner public support for conservation efforts and efforts directed at selecting, preparing and protecting habitat for the

6 56 P. J. SEDDON ET AL. Table 5. Taxonomic bias in reintroduction projects within Orders of the Class Aves (following Howard & Moore, 1980) Number of species % of species Number of projects Number of projects Absolute Pearson described described (observed) (expected) residual residual Anseriformes Apodiformes Apterygiformes Caprimugiformes Casuariformes Charadriiformes Ciconiiformes Coliiformes Columbiformes Coraciiformes Cuculiformes Falconiformes Galliformes Gaviiformes Gruiformes Passeriformes Pelecaniformes Piciformes Podicipediformes Procellariformes Psittaciformes Rheiformes Sphenisciformes Strigiformes Struthioniformes Tinamiformes Trogoniformes Totals Column headings as for Table 1. Positive residuals indicate groups that are over-represented within reintroduction projects, while negative residuals indicate underrepresentation. Values in bold indicate a moderate degree of bias. restoration of populations of mammals and birds may incidentally enable the persistence of populations of rare and threatened species in other and in lower order taxa. These two wider benefits are encompassed within the concept of surrogate species, whereby a focus on one or a small number of species is used to address more general conservation problems (Caro & O Doherty, 1999). Species chosen for reintroduction could act as surrogates by being either flagship species and/or umbrella species (Simberloff, 1998; Caro & O Doherty, 1999). Flagships are popular charismatic species that serve as symbols to stimulate conservation awareness and action locally, nationally, or even globally in the case of species such as the giant panda (Dietz, Dietz & Nagagata, 1994). Umbrella species are those with sufficiently large area requirements that protection of viable populations will bring other species under protection. However, the choice of surrogate species has tended to be ad hoc (Andelman & Fagan, 2000) and can fail to consider adequately the perceptions of local communities (Bowden-Jones & Entwistle, 2002). Also, recent studies have found little evidence to support the claim that a focus on surrogate species is an effective means of conserving biodiversity at regional (Andelman & Fagan, 2000; Williams, Burgess & Rahbek, 2000) or even local (Caro et al., 2004) scales. Species that are to be the focus of reintroduction programmes will be strategically chosen from a potentially large set of candidates based on a number of possibly competing interests, the degree of threat or risk of total extinction being only one. Where species are being reintroduced at least in part due to a possible role as surrogates, this should be explicitly acknowledged since the goals and selection criteria will differ depending on whether a species is considered to be a flagship or an umbrella (Caro & O Doherty, 1999). Furthermore, the assumed benefits of intensive conservation efforts centred on a single species should be specifically examined. The key will be to ensure that our infatuation with charismatic fauna does not divert scarce conservation resources away from species more in need of attention.

7 Table 6. Relative vulnerability to threat among Orders of the class Aves. Taxonomic bias in reintroduction projects 57 Number of species Deviation from described Number assessed Number threatened % threatened mean Ranking Anseriformes Apodiformes Apterygiformes Caprimugiformes Casuariformes = Charadriiformes Ciconiiformes Coliiformes = Columbiformes Coraciiformes Cuculiformes Falconiformes = Galliformes Gaviiformes = Gruiformes Passeriformes = Pelecaniformes Piciformes Podicipediformes Procellariformes Psittaciformes Rheiformes = Sphenisciformes Strigiformes Struthioniformes = Tinamiformes = Trogoniformes = Total mean = 19 Number threatened is the number of species in each Order classified within the IUCN Red List (IUCN, 2003) as Extinct, Critically Endangered, Endangered, or Vulnerable. Number of species assessed excludes those listed as Data Deficient in the IUCN (2003) listing, but includes those listed as Least Concern (derived from BirdLife, 2004). The deviation from the mean is the difference between the proportion of a given Order listed as threatened and the overall mean proportion threatened derived from all bird Orders. Positive deviations indicate that a given order contains relatively more threatened species. Values in bold type indicate extreme positive deviations. Table 7. Distribution of IUCN Red List categories among bird reintroduction projects (this study; BirdLife, 2004) compared with the distribution of categories for all assessed bird species (IUCN, 2003) IUCN Red List Number of bird % of bird % of all category projects projects assessed birds EW < 0.1 CR EN VU NT LC Totals EW, extinct in the wild; CR, critically endangered; EN, endangered; Vu, vulnerable; NT, near threatened; LC, least concern. See IUCN (2003) for category definitions. Acknowledgements We thank Ian Jamieson, Richard Maloney and Yolanda van Heezik for comments on earlier drafts of the manuscript. The assessment of taxonomic bias was improved by suggestions from two anonymous referees. We are grateful to all the IUCN/SSC Reintroduction Specialist Group chairs for assistance in compiling a comprehensive list of reintroduction projects. Stuart Butchart and Sue Shutes kindly facilitated access to Birdlife databases. REFERENCES Andelman, S. J. & Fagan, W. F. (2000). Umbrellas and flagships: efficient conservation surrogates or expensive mistakes? Proc. Natl Acad. Sci. USA 97: BirdLife. (2004). Threatened birds of the world World Bird Database. Cambridge: BirdLife International.

8 58 P. J. SEDDON ET AL. Bonnet, X., Shine, R. & Lourdais, O. (2002). Taxonomic chauvinism. Trends Ecol. Evol. 17: 1 3. Bowen-Jones, E. & Entwistle, A. (2002). Identifying appropriate flagship species: the importance of culture and local contexts. Oryx 36: Caro, T. M. & O Doherty, G. O. (1999). On the use of surrogate species in conservation biology. Conserv. Biol. 13: Caro, T. M., Englilis Jr, A., Fitzherbert, E. & Gardner, T. (2004). Preliminary assessment of the flagship species concept at a small scale. Anim. Conserv. 7: Clark, J. A. & May, R. (2002a). Taxonomic bias in conservation research. Science 297: 191. Clark, J. A. & May, R. (2002b). How biased are we? Conserv. Pract. 3: Czech, B., Krausman, P. R. & Borkhataria, R. (1998). Social construction, political power, and the allocation of benefits to endangered species. Conserv. Biol. 12: Dietz, J. M., Dietz, L. A. & Nagagata, E. Y. (1994). The effective use of flagship species for conservation of biodiversity: the example of lion tamarins in Brazil. In Creative conservation: Olney, P. J. S., Mace, G. M. & Feistner, A. T. C. (Eds). London: Chapman & Hall. Gaston, K. & May, R. (1992). Taxonomy of taxonomists. Nature 356: Howard, R. & Moore, A. (1984). A complete checklist of the birds of the world. London: Macmillan Publishers Ltd. IUCN. (1998). IUCN guidelines for re-introductions. Gland, Switzerland and Cambridge: IUCN. IUCN. (2003). IUCN red list of threatened species (Retrieved from: May 2004). Pawar, S. (2003). Taxonomic chauvinism and the methodologically challenged. BioScience 53: Quinn, G. P. & Keough, M. J. (2002). Experimental design and data analysis for biologists. Cambridge: Cambridge University Press. Simberloff, D. (1998). Flagships, umbrellas, and keystones: is singlespecies management passé in the landscape era? Biol. Conserv. 83: Stein, B. A., Master, L. L. & Morse, L. E. (2002). Taxonomic bias and vulnerable species. Science 297: Williams, P. H., Burgess, N. D. & Rahbek, C. (2000). Flagship species, ecological complementarity and conserving the diversity of mammals and birds in sub-saharan Africa. Anim. Conserv. 3: