sign in sign up
<<
Home , Charismatic megafauna, Conservation status, Flagship species, Mycteria, Neophema

Macdonald/Key Topics in 1405122498_4_002 Final Proof page 17 6.5.2006 11:07pm

2 Prioritizing choices in conservation

______Georgina M. Mace, Hugh P. Possingham and Nigel Leader-Williams The last word in ignorance is the man who says of an or plant: ‘What good is it?’ If the land mechanism as a whole is good, then every part is good, whether we understand it or not. If the biota, in the course of aeons, has built something we like but do not understand, then who but a fool would discard seemingly useless parts? To keep every cog and wheel is the first precaution of intelligent tinkering. (Aldo Leopold, Round River, Oxford University Press, New York, 1993, pp. 145–6.)

tice in other sectors. In business, many com- Introduction panies spend about 10% of the value of their capital assets each year on maintaining those assets, although the figure varies depending on We are in the midst of a mass in the type of asset. For example, 30% might be which at least 10%, and may be as much as spent for computers compared with 5% for 50%, of the world’s may disappear buildings: contrast that with 0.02% for bio- over the next few hundred years. Conservation diversity! Furthermore, the scale of underin- practitioners face the dilemma that the cost of vestment in biodiversity may be exaggerated maintaining global biodiversity far exceeds the by the effects of poor governance, sometimes available financial and human resources. Esti- even corruption, on achieving success in mates suggest that in the late twentieth century conservation (Smith et al. 2003). Given such only US$6 billion per year was spent globally problems, conservation scientists and non- on protecting biodiversity (James et al. 1999), government organizations (NGOs) supporting even though an estimated US$33 trillion per international conservation efforts are begin- year of direct and indirect benefits were derived ning to develop systems to more effectively from services provided by biodiver- target investment in biodiversity conservation sity, implying an asset worth US$330 trillion (Johnson 1995; Kershaw et al. 1995; Olson & (Costanza et al. 1997). Together these crude Dinerstein 1998; Myers et al. 2000; Possingham estimates suggest that there could be a 500- et al. 2001; Wilson et al. in press). fold underinvestment in conserving the world’s One fundamental allocation question biodiversity. However, even if these estimates facing conservation scientists and practitioners is are wildly wrong, the imbalance of funding is whether conservation goals are best met by man- seriously inconsistent with best business prac- aging single as opposed to whole ecosys- Macdonald/Key Topics in Conservation Biology 1405122498_4_002 Final Proof page 18 6.5.2006 11:07pm

18 G.M. MACE, H.P. POSSINGHAM AND N. LEADER-WILLIAMS tems (Simberloff 1998). Efforts in conservation clude judgments that cannot be made by scien- priority setting have historically concentrated on tists or managers alone. Involving wider ecosystem-based priorities – determining where societal and political decision-making processes and when to acquire protected areas (Ferrier et is vital to gain local support for, and ensure the al. 2000; Margules & Pressey 2000; Pressey & ultimate success of, all conservation planning. Taffs 2001; Meir et al. 2004). There has been comparatively little work on the question of how to allocate conservation effort between Single species approaches species. Despite the tension between ecosystem- based and species-based conservation, we believe there is merit in considering the issue of resource Species-based conservation management ap- allocation between species because: proaches have, until fairly recently, concentrated on a single species, such as , 1. a ‘fuzzy’ idea such as ecosystem manage- , indicator species or flagship ment holds little appeal for the general pub- species (see Leader-Williams & Dublin 2000). lic, who prefer to grasp simpler messages conveyed by charismatic species such as Keystone and umbrella species differ in the im- (Leader-Williams & Dublin 2000); portance of their ecological role in an ecosystem: 2. data on species, whether through direct counts of indicator species (Heywood 1. keystone species have a disproportionate 1995), or through assessments of threat effect on their ecosystem, due to their size or (Butchart et al. 2005), provide some of the activity, and any change in their population most readily available, repeatable and expli- will have correspondingly large effects on cit monitoring and analytic systems with their ecosystem (e.g. the sole fruit disperser which to assess the success or otherwise of of many species of tree); conservation efforts (Balmford et al. 2005). 2. umbrella species have such demanding 3. in practice, almost regardless of their ultim- and/or area requirements that, if we ate goal, conservation bodies often end up can conserve enough land to ensure their directing conservation actions to species viability, the viability of smaller and more and species communities (see e.g. figure 1 abundant species is almost guaranteed. of Redford et al. 2003), probably because these are tangible and manageable compon- In contrast, ‘’ encompass ents of . purely strategic objectives: The topic of setting priorities for conservation is 3. flagship species are chosen strategically to immense, so here we restrict ourselves to dif- raise public awareness or financial support ferent methods for setting priorities between for conservation action. species. We explore the issues that a systematic approach should consider, and we show how Furthermore, definitions for indicator species simple scoring systems may lead to unintended can encompass both ecological and strategic consequences. We also recommend an explicit roles: discussion of attributes of the species that make them desirable targets for conservation effort. 4. indicator species are intended either to Using a case study, we show how different represent composition or to re- perspectives will affect the outcome, and so as flect environmental change. With respect to an alternative we present a method based the latter, indicator species must respond to on economic optimization. Ultimately, any the particular environmental change of con- decisions about ‘what to save first’ should in- Macdonald/Key Topics in Conservation Biology 1405122498_4_002 Final Proof page 19 6.5.2006 11:07pm

PRIORITIZING CHOICES IN CONSERVATION 19

cern and demonstrate that change when predators have a different view (Leader- monitored. Williams & Dublin 2000). Such dissonance is best avoided by promoting locally supported One species may be a priority species for flagship species (Entwistle 2000). For example, more than one reason, depending on the situ- the discovery of a new species of an uncharis- ation or context in which the term is used. matic, but virus-resistant, wild maize, with its However, the terms ‘keystone’ and ‘umbrella’ possible utilitarian value for human food pro- are likely to remain more of a fixed character- duction, highlighted the conservation value of istic or property of that species. In contrast, the the Mexican mountains in which it was found term ‘flagship’ and, possibly to a lesser extent, (Iltis 1988). This increase in local public aware- ‘indicator’ may be more context-specific. ness led to the establishment of a Promoting the conservation of a specific focal that conserves and (Panthera species may help to identify potential areas for onca), orchids and (Leopardus pardalis), conservation that satisfy the needs of other spe- species that many consider charismatic. Hence cies and species assemblages (Leader-Williams this species of wild maize served as a strategic- & Dublin 2000). For example, the umbrella ally astute local flagship species. Another way species concept (Simberloff 1998) can represent of promoting local flagships is to prioritize those an efficient first step to protect other species. In species that bring significant and obvious local addition, minimizing the number of species benefits (Goodwin & Leader-Williams 2000), that must be monitored once a protected area such as the , Varanus komodoen- has been created will reduce the time and sis (Walpole & Leader-Williams 2001), which money that must be devoted to its maintenance generates tourism. Similarly, species that can (Berger 1997). be hunted for sport, such as the African ele- Alternatively, conservation managers and phants (Loxodonta africana), may contribute dir- international NGOs may choose to focus on ectly to community conservation programmes the most charismatic ‘flagship’ species, which (Bond 1994). stimulate public support for conservation ac- Several questions can arise from promoting tion, and that in turn may have spin-off bene- conservation through single species (Simberloff fits for other species. For example, use of the 1998). One of these is how should individual (Ailuropoda melanoleuca) as a logo species be prioritized? The common response is by the World Wildlife Fund (WWF) has been to begin with species that are most at risk of widely accepted (Dietz et al. 1994) as a success- extinction, the critically . ful mechanism for conserving many other spe- Many countries and agencies take this ap- cies across a wide variety of taxonomic groups. proach. However, there may be no known Furthermore, other mammalian and avian management for some of these species, and if ‘flagships’ have been used to promote the con- there is, it may be risky and/or expensive. This servation of large natural ecosystems (i.e. Mit- can lead to a large share of limited conservation termeier 1986; Goldspink et al. 1998; Downer resources being expended with negligible or 1996; Johnsingh & Joshua 1994; Western uncertain benefit (Possingham et al. 2002). 1987; Dietz et al. 1994). On the other hand, when taking an ecosystem Nevertheless, the context of what may con- approach, managers might choose to focus on stitute a charismatic species can differ widely the keystone species that play the most signifi- across stakeholders. For example, the cant role in the ecosystem. Unfortunately in (Panthera tigris) is among the most popular flag- many ecosystems we do not know the identity ship species in developed countries, but those of keystone species. Often, after intensive in developing countries who suffer loss of life study, they turn out to be invertebrates or and livelihood because of tigers or other large fungi (Paine 1995), groups that are unlikely to Macdonald/Key Topics in Conservation Biology 1405122498_4_002 Final Proof page 20 6.5.2006 11:07pm

20 G.M. MACE, H.P. POSSINGHAM AND N. LEADER-WILLIAMS attract public or government support unless cies, it does highlight the need for more field- ways can be found to make them locally based studies to determine the most appropriate relevant. approach for conserving the most biodiversity. Another problem with single species conser- As a result of problems associated with single vation arises when the management of one focal species management, focus has been turning species is detrimental to the management of towards multiple species approaches. another focal species. For example, in the Ever- glades of Florida, management plans for two charismatic, federally listed are in conflict. Multispecies approaches One species, the Everglades (Rostrha- mus sociabilis plumbeus), has been reduced to some 600 individuals by wetland degradation Methods based on several focal species, or pro- and agricultural and residential development. tecting a specific habitat type, might be a more It feeds almost exclusively on freshwater snails appropriate means of prioritizing protected of the Pomacea and requires high water areas (Lambeck 1997; Fleishman et al. 2000; levels, which increase snail production. The Sanderson et al. 2002b). A frequent criticism of snail kite is thus an extreme habitat specialist setting conservation priorities based on a single (Ehrlich et al. 1992). The other species, the focal species is that it is improbable that the wood Mycteria americana, has been reduced requirements of one species would encapsulate to about 10,000 pairs by swamp drainage, habi- those of all other species (Noss et al. 1996; Basset tat modification and altered water regimes. et al. 2000; Hess & King 2002; Lindenmayer et al. Ironically, the US and Wildlife Service op- 2002). Hence, there is a need for multispecies posed a proposal by the Everglades National strategies to broaden the coverage of the pro- Park to modify water flow to improve stork tective umbrella (e.g. Miller et al. 1999; Fleish- habitat on the grounds that the change would man et al. 2000, 2001; Carroll et al. 2001). be detrimental to the kite (Ehrlich et al. 1992) Among the different multispecies ap- (an added thought-provoking detail is that both proaches, Lambeck’s (1997) ‘focal species’ ap- species are common in South America). proach seems the most promising because it Another issue is that few studies have been provides a systematic procedure for selecting carried out to assess the effectiveness of one several focal species (Lambeck 1997; Watson focal species in adequately protecting viable et al. 2001; Bani et al. 2002; Brooker 2002; populations of other species (Caro et al. 2004). Hess & King 2002). In Lambeck’s (1997) in- For example, the umbrella-species concept is novative approach, a suite of focal species are often applied in management yet rarely tested. identified and used to define the spatial, com- The grizzly bear (Ursus arctos) has been recog- positional and functional attributes that must nized as an umbrella species but, had a pro- be present in a landscape. The process involves posed conservation plan for the grizzly bear in identifying the main threats to biodiversity and Idaho been implemented, taxa such as reptiles selecting the species that is most sensitive to would have been underrepresented (Noss et al. each threat. The requirements of this small 1996). Similarly, in a smaller scale study, the and manageable suite of focal species guide areas where flagship species, such as , conservation actions. The approach was tapir (Tapirus terrestris) and white-lipped pec- extended by Sanderson et al. (2002a), who cary (Tayassu pecari), were most commonly proposed the ‘landscape species approach’. seen did not coincide with areas of vertebrate They defined landscape species by their ‘use of or (Caro et al. 2004). large, ecologically diverse areas and their im- Although these results may not hold true for all pacts on the structure and function of natural other protected areas based around flagship spe- ecosystems . . . their requirements in time and Macdonald/Key Topics in Conservation Biology 1405122498_4_002 Final Proof page 21 6.5.2006 11:07pm

PRIORITIZING CHOICES IN CONSERVATION 21 space make them particularly susceptible to 4. manage over periods of time long enough human alteration and use of wild landscapes’. to maintain the evolutionary potential of Because landscape species require large, wild species; areas, they could potentially serve an umbrella 5. accommodate human use and occupancy function (sensu Caro & O’Doherty 1999) – within these constraints (Grumbine 1994). meeting their needs would provide substantial Although the financial efficiencies inherent protection for the species with which they co- in managing an ecosystem rather than several occur. Like other focal-species approaches, this single species are attractive, this approach is method of setting priorities carries inherent also not without its problems. First, compared biases (Lindenmayer et al. 2002), and may be with a species, ecosystem boundaries are constrained by incomplete or inconsistent data. harder to define, so determining the location, size, connectivity and spacing of protected areas Ecosystem and habitat-based to conserve the full range of ecosystems, and approaches variation within those ecosystems, is more dif- ficult (Possingham et al. 2005). Second, indi- vidual species are more interesting to people and will attract greater emotional and financial Some conservation scientists believe that set- ting conservation priorities at the scale of eco- investments than ecosystems. Third, although ecological services are provided by ecosystems, systems and is more appropriate for individual species often play pivotal roles in the developing countries with limited resources for conservation, inadequate information about provision of these services, particularly for dir- ect uses such as tourism or harvesting. Finally, single species and pressing threats such as the main problem faced by managers wishing . Logically, how much effort we place in conserving a particular ecosystem to implement an ecosystem approach is the lack of data available on how ecosystems function. should take into account factors such as: how This manifests itself in confusion about how threatened it is, how well represented that ecosystem is in that country’s protected area much of each ecosystem needs to be conserved to protect biodiversity adequately in a region. network, the number of species restricted to In contrast, for the better known single species, that ecosystem (endemic species), the cost of conserving the ecosystem and the likelihood the issue of adequacy can be dealt with using population viability analysis and/or harvesting that conservation actions will work. One can models (Beissinger & Westphal 1998; this vol- debate the relative importance of each of these factors – for example, some consider the the ume, Chapter 15). number of endemic species is paramount, whereas others prefer the notion of ‘equal rep- resentation’ whereby a fixed percentage of Systematic conservation planning every habitat type is conserved. The main goals of an ecosystem approach are to: Systematic conservation planning (or gap-an- alysis in the USA: Scott et al. 1993) focuses on 1. maintain viable populations of all native locating and designing protected areas that species in situ; comprehensively represent the biodiversity of 2. represent, within protected areas, all native each region. Without a systematic approach, ecosystem types across their natural range protected area networks have the tendency to of variation; occur in economically unproductive areas 3. maintain evolutionary and ecological processes; (Leader-Williams et al. 1990), leaving many Macdonald/Key Topics in Conservation Biology 1405122498_4_002 Final Proof page 22 6.5.2006 11:07pm

22 G.M. MACE, H.P. POSSINGHAM AND N. LEADER-WILLIAMS habitats or ecosystems with little or no protec- threats. The degree to which different countries tion (Pressey 1994). The systematic conserva- use species-based planning as opposed to sys- tion planning approach can be divided into six tematic conservation planning depends on his- stages (Margules & Pressey 2000). torical, cultural and legislative influences. Even with systematic conservation planning, how- 1. Compile biodiversity data in the region of ever, the better surveyed species or species concern. This includes collating existing groups often feature as the units for assessment. data, along with collecting new data if ne- In other words, the conservation value of dif- cessary, and if time and funds permit. ferent areas is often assessed on the presence or Where biodiversity data, such as status of the species within it, maps and species distributions, are limited simply because these are the best known elem- more readily available biophysical data may ents of biodiversity. Systematic conservation be used that reflect variation in biodiversity, such as mean annual rainfall or soil type. planning approaches have become popular 2. Identify conservation goals for the region, and widespread, partly because they are sup- including setting conservation targets for ported by several decision-support software species and habitats, and principles for pro- packages (Possingham et al 2000, Pressey et al tected area design, such as maximizing con- 1995, Williams et al 2000, Garson et al 2002). nectivity and minimizing the edge-to-area ratio. 3. Review existing conservation areas, includ- ing determining the extent to which they Methods for setting conservation already meet quantitative targets, and miti- priorities of species gate threats. 4. Select additional conservation areas in the region using systematic conservation plan- Prioritizing species, habitats and ecosystems by ning software. their perceived level of endangerment has be- 5. Implement conservation action, including come a standard practice in the field of conser- decisions on the most appropriate form of vation biology (Rabinowitz 1981; Master 1991; management to be applied. Mace & Collar 1995; Carter et al. 2000; Stein 6. Maintain the required values of the conser- vation areas. This includes setting conserva- et al. 2000). The need for a priority-setting tion goals for each area, and monitoring key process is driven by limited conservation re- indicators that will reflect the success of sources that necessitate choices among a subset management (see below). of all possible species in any given geographical area, and distinct differences among species in Ultimately, conservation planning is riddled their apparent vulnerability to extinction or with uncertainty, so managers must learn to need for conservation action. This need has deal explicitly with uncertainty in ways that led to the development of practical systems minimize the chances for major mistakes (Mar- for categorizing and assessing the degree of vul- gules & Pressey 2000; Arau´ jo & Williams 2000, nerability of various components of biodiver- Wilson et al 2005), and be prepared to modify sity, particularly vertebrates (e.g. Millsap et al. their management goals appropriately through 1990; Mace & Lande 1991; Master 1991; Reed adaptive management. 1992; Stotz et al. 1996), and more recently Systematic conservation planning can com- ecoregions (Hoekstra et al. 2005). plement species-based approaches because it Methods used for assessing the conservation focuses on removing the threat of development status of species are varied but follow three and it compliments a long tradition of species general styles (Regan et al. 2004), rule-based, recovery plans that concentrate on mitigating point scoring and qualitative judgement. Per- Macdonald/Key Topics in Conservation Biology 1405122498_4_002 Final Proof page 23 6.5.2006 11:07pm

PRIORITIZING CHOICES IN CONSERVATION 23 haps the best known system is that developed more of the criteria – for example because there by the IUCN (International Union for the Con- are less than 250 mature individuals of the servation of Nature and Natural Resources) – Norfolk Island green (Cyanoramphus coo- The World Conservation Union – which uses a kii) in the wild it is immediately listed as endan- set of five quantitative rules with explicit gered under criterion D of the IUCN Red List thresholds to assign a risk of extinction (Mace protocol. A similar species can meet a higher & Lande 1991; IUCN 2001). Other methods category of threat if it meets alternative cri- adopt point-scoring approaches where points teria. For example, the orange-bellied parrot are assigned for a number of attributes and (Neophema chrysogaster) also has less than 250 summed to indicate conservation priority (Mill- mature individuals but it is listed as critically sap et al. 1990; Lunney et al. 1996; Carter et al. endangered, under criterion C2b, because the 2000). Other methods assess conservation sta- population is also in decline and all the individ- tus using qualitative criteria; judgements about uals are in a single subpopulation. One concep- a species’ status are determined intuitively tual problem with rule-based methods is that a based on available information and expert species that just missed out on being listed as, opinion (Master 1991). One widely applied sys- say, endangered on several criteria would be tem is the biodiversity status-ranking system ranked as vulnerable, equal with a species that developed and used by the Natural Heritage may have only just met the criteria for being Network and (Master vulnerable. 1991; Morse 1993). This ranking system has The rule-based methods have the advantage been designed to evaluate the biological and that they are completely explicit about what conservation status of plant and animal species feature of the species led to it being listed as and within-species taxa, as well as of ecological threatened. In the IUCN system, assessors have communities. to list the criteria whereby the species qualified for a particular category of threat, and also have to provide documentation to support this infor- Rule-based methods mation – usually in the form of scientific sur- veys or field reports that detail the information Quantitative rule-based methods can be used to used. As a result, listings may be continually estimate the extinction risk of a species and updated and improved as new data become thus contribute to determining priority areas available. Normally this will allow a new con- for conservation action. For example, the sensus among experts, but in the exceptional IUCN Red List places species in one of the fol- cases where this is not agreed, the IUCN have a lowing categories: extinct (EX), extinct in the petitions and appeals process to resolve matters. wild (EW), (CR), endan- For example, in 2001 some of the listings of gered (EN), vulnerable (V), near threatened marine turtle species were disputed among ex- (NT) or least concern (LC), based on quantita- perts. On this occasion, IUCN implemented tive information for known life history, habitat their appeal procedure and provided a new requirements, abundance, distribution, threats assessment (http://www.iucn.org/themes/ssc/ and any specified management options of that redlists/petitions.html). The wide use of the species, and in a (DD) category if IUCN system also means that there is an ever there are insufficient data to make an assess- increasing resource of best-practice documen- ment (IUCN 2001). The IUCN system is based tation and guidelines, which aid consistent and around five criteria (A to E) which reflect dif- comparable approaches by different species ferent ways in which a species might qualify for assessors (see http://www.iucn.org/themes/ssc/red- any of the threat categories (CR, EN, VU). A lists.htm). species is placed in a category if it meets one or Macdonald/Key Topics in Conservation Biology 1405122498_4_002 Final Proof page 24 6.5.2006 11:07pm

24 G.M. MACE, H.P. POSSINGHAM AND N. LEADER-WILLIAMS

A problem with some point-scoring methods Point scoring method is that there is no explicit link to extinction risk, the weightings of each criteria, from 1 to 5 in The point scoring method for assigning conser- the example above, are completely arbitrary, vation priority involves assigning a series of and there is an infinity of ways in which the scores to each species based on different param- scores could be combined: adding, multiplying, eters relating to their or conservation taking the product of the largest three values, status, which together will determine their and so forth. A related problem is that point- relative priority. One method of dealing with scoring methods can generate an artificially the scores is then to simply sum them to give an high ranking for a species when criteria are overall conservation priority, although this can interrelated. For example, a system that priori- be misleading. Beissinger et al. (2000) suggest tized species because they needed large home that a categorical approach based on a combin- ranges, had slow reproductive rates and small ation of scores might be more accurate in litter sizes might end up allocating unreason- determining overall conservation priority. ably high scores to any large-bodied species. An example of a point scoring system is that All three of these traits are associated with developed by Partners in Flight (PIF) in 1995 in relatively large body size, but they are not an effort to conserve non-game birds and their necessarily so much more vulnerable. habitats throughout the USA (Carter et al. 2000). The PIF system involves assigning a series of scores to each species ranging from 1 (low priority) to 5 (high priority) for seven param- Conservation status ranks method eters that reflect different degrees of need for conservation attention. The scores are assigned Status ranks are based primarily on objective within physiographical areas and the seven factors relating to a species’ rarity, population parameters are based on global and local infor- trends and threats. Four aspects of rarity are mation. Three of the parameters are strictly typically considered: the number of individuals, global and are assigned for the entire range of number of populations or occurrences, rarity of the : breeding distribution (BD), non-breed- habitat, and size of geographic range. Ranking ing distribution (ND) and relative abundance is based on an approximately logarithmic scale, (AR). Other parameters are threats to breeding ranging from 1 (critically imperiled) to 5 (dem- (TB), threats to non-breeding (TN), population onstrably secure). Typically species with ranks trend (PT) and, locally, area importance (AI). from 1 to 3 would be considered of conserva- The scores for each of these seven parameters tion concern and broadly overlap with species are obtained independently (Carter et al. 2000). that might be considered for review under the The PIF then uses a combination of approaches, Endangered Species Act or similar state or including the summing of scores, to determine international statutes. an overall conservation priority (Carter et al. The NatureServe system (Master 1991) is one 2000), with species that score highly on several example of a system that uses status ranks. parameters achieving high priority. Although Developed initially by The Nature Conservancy this method of defining bird species of high con- (TNC) and applied throughout North America, servation priority is thought to be reliable, like the NatureServe system uses trained experts other methodologies, it is hindered by the lack who evaluate quantitative data and make in- of data on , abundance and tuitive judgements about species vulnerability. populations trends, particularly in areas outside The aim of the NatureServe system is to deter- the USA to which many of these species migrate mine the relative susceptibility of a species or (Carter et al. 2000). ecological community to extinction or extirpa- Macdonald/Key Topics in Conservation Biology 1405122498_4_002 Final Proof page 25 6.5.2006 11:07pm

PRIORITIZING CHOICES IN CONSERVATION 25 tion. To achieve this, assessments consider both question of how quickly action needs to be deterministic and stochastic processes that can taken. Hence, all other things being equal, the lead to extinction. Deterministic factors include critically endangered species will be most likely habitat destruction or alteration, non-indigen- to become extinct first if nothing is done. How- ous predators, competitors, or parasites, over- ever, this is by no means the only consideration harvesting and environmental shifts such as that should be used by a conservation planner. . Stochastic factors include, en- How then should extinction risk be used for vironmental and demographic stochasticity, priority-setting? It may be easier to make the natural catastrophes and genetic effects (Shaffer analogy with a different system altogether. For 1981). example, the priority-setting systems used by NatureServe assessments are performed on a Triage nurses in hospital emergency depart- basic unit called an element. An element can ments categorize people according to how ur- be any plant or animal species or infraspecific gently they need to be seen; those seen first are taxon (subspecies or variety), ecological com- the ones that appear to have the most urgent munity, or other non-taxonomic biological and threatening symptoms. The symptoms can entity, such as a distinctive population (e.g. be very diverse, however, and some may turn evolutionarily significant unit or distinct popu- out upon inspection and diagnosis to be less lation segment, as defined by some agencies) or serious than might have been expected. Medical a consistently occurring mixed species aggrega- planning across the board would not use the tion of migratory species (e.g. shorebird migra- triage system to allocate resources. The same is tory concentration area) (Regan et al. 2004). true for conservation planning. As with ill and Defining elements in this way ensures that a injured people, our first sorting of cases should broad spectrum of biodiversity and ecological be according to urgency, and should also be processes are identified and targeted for conser- precautionary (i.e. take more risks with listing vation (Stein et al. 2000). This approach is be- species that are in fact not threatened than with lieved to be an efficient and effective approach failing to list those that really are). However, to capturing biodiversity in a network of once the diagnosis is made, and the manager reserves (e.g. Jenkins 1976, 1996). Assessment is reasonably sure that most critical cases are results in a numeric code or rank that reflects now known and diagnosed, a more systematic an element’s relative degree of imperilment or planning process should follow. risk of extinction at either the global, national or subnational scale (Master et al. 2000). Variables other than risk

Back to basics – extinction risk versus setting priorities Now we consider a whole range of new variables other than risk. Table 2.1 shows a range of vari- The discussion above has reviewed methods for ables – grouped under headings of biological categorizing species according to their conser- value (i.e. what biologists would consider), eco- vation priority. Running throughout is a ten- nomic value, social and cultural value, urgency dency to equate conservation priority with and practical issues. Under each of these head- extinction risk; yet these are clearly not the ings are a range of attributes that might contrib- same thing (Mace & Lande 1991). Extinction ute to a species priority. The first three columns risk is only one of a range of considerations that concern values, but the last two are rather dif- determine priorities for action or for conserva- ferent. Urgency is a measure that can be compli- tion funding. The threat assessment is really an cated to implement – i.e. high urgency may assessment of urgency, and an answer to the indicate that if nothing is done now, then it Macdonald/Key Topics in Conservation Biology 1405122498_4_002 Final Proof page 26 6.5.2006 11:07pm

26 G.M. MACE, H.P. POSSINGHAM AND N. LEADER-WILLIAMS

Table 2.1 Classes and kinds of issues that are considered in priority-setting exercises for single-species recovery

Social and Biological value Economic value cultural value Urgency Practical issues

Degree of Cost of management Scientific and Threat status Feasibility and logistics or recovery educational benefits ¼ extinction risk Relictual status Direct economic Cultural status Time limitation, Recoverability, i.e. benefits (e.g. ceremonial) i.e. opportunities reversibility of threats, will be lost later rate of species response Evolutionary Indirect economic Political status Timeliness, i.e. Popularity – will there uniqueness benefit (e.g. symbolic or likelihood of be support from the emblematic) success varies community? with time Collateral benefits to Ecological services Popularity Responsibility, i.e. how other species much is this also someone else’s responsibility? Collateral costs to Local or regional Land tenure other species significance Ecological uniqueness Governmental/agency jurisdictions Keystone species status Umbrella species status

will be too late. This measure is not a value score Interestingly, the criteria that biologists com- that can easily be added to the others, and a monly consider, and which form the basis of moderate score has little meaning. Practical most formal decision-processes, fall under one issues are also rather different, and will vary heading (biological benefits). Yet in practice, greatly in their nature and importance depend- the other five criteria (Table 2.2) also influence ing on the context. Some species that are con- real decisions. Would it not, therefore, be pref- sidered urgent cases may be extremely erable to incorporate these other criteria expli- impractical and/or costly to attend to. This set citly in the process of setting priorities? of considerations is probably not complete, but it does illustrate the point that there are more things to think about than extinction risk. Turning criterion-based ranks This initial classification by the value type is into priorities hard to manage in a priority-setting system. Therefore, in Table 2.2, we classify these into A potential next step would be to add the scores six criteria reflecting the nature of the attribute from Table 2.2. By allocating a score of 1, 2 or 3 (importance, feasibility, biological benefits, to each criterion and then adding the ranks, an economic benefits, urgency and chance of suc- overall priority could be calculated. We advise cess). This classification has the advantage that against this for several reasons. First, the differ- the different questions are more or less inde- ent variables are not equal; we might for ex- pendent of one another, and each addresses a ample wish to weight the biological issues question that public, policy-makers and scien- more highly. Second, they are not additive: as tists can all address, and for which they can mentioned earlier both urgency and chance provide at least relative scores. of success are all or nothing decisions. For Macdonald/Key Topics in Conservation Biology 1405122498_4_002 Final Proof page 27 6.5.2006 11:07pm

PRIORITIZING CHOICES IN CONSERVATION 27

Table 2.2 Criteria for setting prorities. The different kinds of considerations from Table 2.1 are classified into six criteria (rows), each of which can be qualitatively assessed for a particular species

Criterion Explanation Subcriteria Scores

Importance ‘Does anyone care?’ A Social and cultural importance Important (I) measure of how much (including charisma) Moderately important (M) support there is likely to be Responsibility – Unimportant (U) how much of the species status depends on this project? Feasibility ‘How easy is this to achieve?’ Logistical and political, source of Feasible (F) An assessment of the difficulty funds, community attitudes Moderately difficult (M) associated with this project Biological Difficult (D) Benefits ‘What good will it do?’ A Reduction in extinction risk, Highly beneficial (H) measure of how much good increase in , extent Moderately beneficial (M) will result from the project. of occurrence Unclear benefits (U) Collateral biological benefits, to other species or processes Costs ‘What will it cost?’ An assessment Direct and indirect costs of project Expensive of the relative economic costs Direct and indirect social and Moderately costly of the project (or gains). In this economic costs and benefits that will Inexpensive criterion there are both postive flow from the project and negative aspects which have to be weighed against each other Urgency ‘Can it be delayed?’ A measure Extinction risk, potential for loss of Urgent of whether the project is time- opportunity if delayed Moderately urgent limited, or whether it can be Less urgent delayed Chance of ‘Will it work?’ An assessment of Will it meet its specified objectives? Achievable success whether or not the project will Uncertain work Highly uncertain

example, if chance of success is nil we would may lead to ambiguous results. An expedient not wish to invest in that species at all, so it process at this stage is to invite a range of experts would seem more logical to multiply other representing different perspectives to rate the scores by the chance of success. Third, although priorities explicitly. For example, given the pos- we have sorted the issues into more-or-less sible set of scores in Table 2.2, what set would independent categories, there still are associ- they most wish to see in the top priorities versus ations between them. For example, the feasibil- those lower down? This sounds complicated but ity and chance of success are likely to be in practice we think it is feasible. positively correlated, as are biological benefits A good example of this approach was devel- and importance. Hence, simple scoring can lead oped for UK birds by the Royal Society for the to double-counting, which is not what was in- Protection of Birds (Avery et al. 1995). Three tended. criteria were used: global threat, national de- Multicriteria decision-analysis is one decision- cline rates and national responsibility, and each making tool for choosing between priorities that was rated high, medium or low. However, by rate differently for separate criteria. There are simply adding these scores, globally endan- innumerable ways of carrying out a multicriteria gered species that are stable, and for which analysis, and the process can be complex and the UK has medium responsibility, had the Macdonald/Key Topics in Conservation Biology 1405122498_4_002 Final Proof page 28 6.5.2006 11:07pm

28 G.M. MACE, H.P. POSSINGHAM AND N. LEADER-WILLIAMS

Conservation 1 ance and economic benefits and minimize risk priority 1 of failure. Hence the two profiles would look Global threat set 1 1 11 1 quite different (Fig. 2.2). Figure 2.2 illustrates 2 the different approaches – see how you would 2 222 1 score the criteria in Table 2.2 to make your own 2 set! 2 3 3 2 Here we are effectively creating a complex Responsibility rule set that maps any species into one of three categories without adding or multiplying National decline the scores for different criteria. The method Fig. 2.1 The conservation cube. (From Avery et al. suffers from its somewhat arbitrary nature. 1995.) Below we suggest that optimal allocation of funds between species can be achieved more same priority as globally secure species exhibit- rigorously if we place the problem within an ing slow decline in the UK. This would not be explicit framework in which we can apply most people’s choice; whatever their status in decision theory. the UK, a globally endangered species probably should be in the category of highest priority. Hence, Avery et al. (1995) set priorities using A decision theory approach – optimal their conservation cube (Fig. 2.1). Here they allocation evaluated each of the 27 possible circumstances into three categories for priority. In their A major problem with using scores or ranks for system, any globally and threatened species to determine funding and any species declining at a high rate nationally action priorities is that these methods were not are the highest priority. designed for that task – they were designed to This approach can be taken more generally determine the relative level of threat to a suite of using the six criteria in Table 2.2. By asking species (Possingham et al. 2002). Hence they what would be the criteria associated with top cannot provide the solution to the problem of priority species, it is possible to assemble a pro- optimal resource allocation between species – file. For example, whereas a species conserva- this problem should be formulated then solved tion ‘idealist’ might choose to ignore properly (Possingham et al. 2001). importance, feasibility, economic benefits and Optimal allocation is one simple and attract- chance of success, and to focus just on the most ive approach to prioritization that could inform urgent and most threatened forms, a more ‘pol- decisions about how to allocate resources be- itical’ approach would be to maximize import- tween species. It requires information about

Manager 1 Manager 2 Idealist Politician HML HML HML HML Importance Feasibility Biological benefits Economic benefits Urgency Chance of success

Fig. 2.2 Priority sets for four different people. The blocked out cells indicate the conditions under which assessors would choose to include species in their priority set, according to how they scored on the variables in Table 2.2 as H, high; M, medium; L, low. Macdonald/Key Topics in Conservation Biology 1405122498_4_002 Final Proof page 29 6.5.2006 11:07pm

PRIORITIZING CHOICES IN CONSERVATION 29 the relationship between the resources allo- For poorly known taxa the curves showing cated to the species and the reduction in prob- this relationship would very much be a reflec- ability of extinction. Here we use expert tion of expert opinion, garnered by asking opinion and/or population models to estimate questions about how much it might cost to the relationship between percentage recovery give a particular species a 90% chance of not (measured, for example, in terms of probability becoming extinct in the long term. Given a set of not becoming extinct) and the funds allo- amount of money for a set period in the con- cated to that species. servation budget, the optimal allocation of

100 90 80 70 60 50 Scheme to maximise the total amount of recovery, given $ amount 40 Recovery (%) 30 20 10 0 0 1,000,000 2,000,000 3,000,000 4,000,000 5,000,000 6,000,000 (a) $ spent

Species accumulation

16 14 12 10 8 6 4 2 Expected species recovered 0 0.E+00 1.E+07 2.E+07 3.E+07 4.E+07 5.E+07 6.E+07 7.E+07 8.E+07 9.E+07 1.E+08 (b) $ in budget

Fig. 2.3 Optimal allocation. (a) Three curves show the expected recovery for three different species given certain amounts of investment. If the manager has a specified budget (in this case $1 million), the optimal allocation among species that achieves the greatest total amount of recovery will result if funds are allocated as shown by the vertical dotted lines (see Possingham et al. 2002). (b) Increasing investment leads to gradually increasing numbers of species recovered. Macdonald/Key Topics in Conservation Biology 1405122498_4_002 Final Proof page 30 6.5.2006 11:07pm

30 G.M. MACE, H.P. POSSINGHAM AND N. LEADER-WILLIAMS funds can be determined between species. This When species are rated highly by the criteria occurs when the rate of gain of recovery for the two approaches give similar results, but at each species is equal, such that there is no low criterion scores there can be much variabil- advantage in shifting resources from one spe- ity. Perhaps the only general conclusion here is cies to another (see Fig. 2.3 and Possingham that inevitably the optimal allocation approach et al. 2002). The implicit objective is to maxi- will favour some species that, on the basis of the mize the mean number of recovered species criteria, would not be given high priority. In given a fixed budget and assuming all species practice, sensible management could use both are of equal ‘value’. approaches – the criteria to select high-priority Using the set of species plotted in Fig. 2.3, we sets and the financial algorithm to then maxi- estimate the costs of recovery, and then find mize the benefits from the finite resources avail- the optimal allocation of funds per species. able to conservation. The species accumulation curve shows the total expected number of species that can be recovered given a conservation budget. The al- Conclusions gorithm will tend first to select species that show large recovery for relatively low costs. Slow responders will be conserved later. Given Priority setting needs to consider a range of an annual budget basis, the more intractable variables, and although this undoubtedly oc- conservation problems may never be funded curs, it is not always transparent. Although because the selection process will always favour much effort has gone into biologically based allocation of resources to the species that pro- systems, in practice other societal value judg- vide the greatest gains for the smallest costs ments are often included. We suggest that, if (the low-hanging fruit). conservation goals are to be achieved, it is vital So how would these two approaches: cri- to be explicit about what these are, and to teria-based prioritization and optimal allocation decide upon them in an open and consultative of resources differ in practice? Obviously there manner before choices are made. is no general answer to this, other than a priori Different people and organizations, and differ- we do not expect them to be the same. The ent sectors in society, will make different choices outcome of a small case study, based on real in their value judgments. Approaches to under- species and the expertise of two real conserva- standing these choices are important so we can tion managers is shown in Fig. 2.4. interpret the differences in setting priorities.

4.5 4 3.5 3 2.5 2 1.5

Optimisation rank 1 0.5 0 012345 Criterion rank

Fig. 2.4 Comparison of priority ranks for18 species using the criteria-based method versus optimal allocation of funds. Macdonald/Key Topics in Conservation Biology 1405122498_4_002 Final Proof page 31 6.5.2006 11:07pm

PRIORITIZING CHOICES IN CONSERVATION 31

We recommend using more than one decisions about resource allocation be formu- method to set priorities, and the comparison lated more explicitly in terms of objectives, can be informative. We also recommend that constraints and costs.

For if one link in nature’s chain might be lost, another might be lost, until the whole of things will vanish by piecemeal. (Thomas Jefferson (1743–1826) in Charles Miller, Jefferson and Nature, 1993.)

Brooker L. (2002) The application of focal species References knowledge to landscape design in agricultural lands using the ecological neighbourhood as a template. Landscape and Urban Planning 60: 185– Arau´ jo, M.B. & Williams, P.H. (2000) Selecting areas 210. for species persistence using occurrence data. Bio- Butchart, S.H.M., Stattersfield, A.J., Bennun, L.A., logical Conservation 96: 331–45. et al. (2005) Measuring global trends in the status Avery, M., Gibbons, D.W., Porter, R., Tew, T., of biodiversity: Red List Indices for birds. Public Tucker, G. & Williams, G. (1995) Revising the Library of Science Biology 2: 2294–304. British Red Data List for birds: the biological Carroll, C., Noss, R.F. & Paquet, P.C. (2001) Carni- basis of UK conservation priorities. 137(S1): vores as focal species for conservation planning in 232–9. the Rocky Mountain region. Ecological Applications Balmford, A., Bennun, L., ten Brink, B., et al. (2005) 11: 961–80. Science and the Convention on Biological Diver- Caro, T. & Doherty, G. (1999) On the use of surro- sity’s 2010 target. Science 307: 212–13. gate species in conservation. Conservation Biology Bani, L., Baietto, M., Bottoni, L. & Massa, R. (2002) 13: 805–14. The use of focal species in designing a habitat Caro, T., Engilis, A., Fitzherbert, E. & Gardner, T. network for a lowland area of Lombardy, Italy. (2004) Preliminary assessment of the flagship species Conservation Biology 16: 826–31. concept at a small scale. Animal Conservation 7: 63–70. Basset, Y., Novotny, V., Miller, S.E. & Pyle, R. (2000) Carter, M.F., Hunter, W.C., Pashley, D.N. & Rosen- Quantifying biodiversity: experience with paratax- berg, K.V. (2000) Setting conservation priorities onomists and digital photography in Papua New for landbirds in the : the Partners in Guinea and Guyana. Bioscience 50(10): 899–908. Flight Approach. Auk 117: 541–8. Beissinger, S.R. & Westphal, M.I. (1998) On the use Costanza, R., d’Arge, R., de Groot, R., et al. (1997) of demographic models of population viability in The value of the world’s ecosystem services and endangered species management. Journal of Wild- natural capital. Nature 387: 253–60. life Management 62: 821–41. Dietz, J.M., Dietz, L.A. & Nagagata, E.Y. (1994) The Beissinger, S.R., Reed, J.M., Wunderle, Jr, J.M., effective use of flagship species for conservation of Robinson, S.K. & Finch, D.M. (2000) Report of the biodiversity: the example of tamarins in Bra- AOU Conservation Committee on the Partners in zil. In Creative Conservation: Interactive Management of Flight species prioritization plan. Auk 117:549–61. Wild and Captive (Eds P.J.S. Olney, G.M. Berger, J. (1997) Population constraints associated Mace & A.T.C. Feistner), pp. 32–49. Chapman & with the use of black rhinos as an umbrella species Hall, London. for desert . Conservation Biology 11:69–78. Downer, C. (1996) The , endangered Bond, I. (1994) Importance of elephant sport hunt- ‘flagship’ species of the high Andes. Oryx 30: 45–58. ing to CAMPFIRE revenue in Zimbabwe. TRAFFIC Ehrlich, P.R., Dobkin, D.S. & Wheye, D. (1992) Birds Bulletin 14: 117–19. in Jeopardy: the Imperiled and Extinct Birds of the Macdonald/Key Topics in Conservation Biology 1405122498_4_002 Final Proof page 32 6.5.2006 11:07pm

32 G.M. MACE, H.P. POSSINGHAM AND N. LEADER-WILLIAMS

United States and Canada, including Hawaii and Puerto Union for the Conservation of Nature and Natural Rico. Stanford University Press, Stanford, CA. Resources, Gland, Switzerland. Entwistle, A. (2000) Flagships for the future? Oryx James, A.N., Gaston, K.J. & Balmford, A. (1999) 34: 239. Balancing the earth’s accounts. Nature 401: Ferrier, S., Pressey, R.L. & Barrett, T.W. (2000) 323–4. A new predictor of the irreplaceability of areas Jenkins, R.E. (1976) Maintenance of natural diver- for achieving a conservation goal, its application sity: approach and recommendations. In Transac- to real-world planning, and a research agenda tions of the 41st North American Wildlife and Natural for further refinement. Biological Conservation Resources Conference 21–25 March, Washington, DC 93: 303–25. (Ed. K. Sabol), pp. 441–51. Fleishman, E., Murphy, D.D. & Brussard, P.F. (2000) Jenkins, R.E. (1996) Natural Heritage Data Centre A new method for selection of umbrella species for Network: managing information for managing bio- conservation planning. Ecological Applications 10: diversity. In Biodiversity in Managed Landscapes: The- 569–79. ory and Practice (Eds R.C. Szaro & D.W. Johnston), Fleishman, E., Blair, R.B. & Murphy, D.D. (2001) pp. 176–92. Oxford University Press, New York. Empirical validation of a method for umbrella spe- Johnson, N.C. (1995) Biodiversity in the Balance: Ap- cies selection. Ecological Applications 11: 1489–501. proaches to Setting Geographic Conservation Priorities. Garson, J., Aggarwal, A. & Sarkar, S. (2002) Birds as Biodiversity Support Program, World Wildlife surrogates for biodiveristy: an analysis of a data set Fund, Washington, DC. from southern Que´bec. Journal of Biosciences Johnsingh, A.J.T. & Joshua, J. (1994) Conserving 27:347–60. Rajaji and Corbett National Parks: the elephant as Goldspink, C.R., Holland, R.K., Sweet, G. & Stjerns- a flagship species. Oryx 28: 135–40. redt, R. (1998) A note on the distribution of the Kershaw, M., Mace, G.M. & Williams, P.H. (1995) puku, Kobus vardoni, in Kasanka National Park, Threatened status, rarity and diversity as alterna- Zambia. African Journal of Ecology 36: 23–33. tive selection measures for protected areas: a test Goodwin, H.J. & Leader-Williams, N. (2000) Tourism using Afrotropical antelopes. Conservation Biology and protected areas: distorting conservation prior- 9: 324–34. ities towards charismatic ? In Priorities Lambeck, R.J. (1997) Focal species: a multi-species for the Conservation of Mammalian Diversity: Has umbrella for . Conservation the Panda had its Day? (Eds A. Entwistle & Biology 11: 849–56. N. Dunstone), pp. 257–75. Cambridge University Leader-Williams, N. & Dublin, H.T. (2000) Charis- Press, Cambridge. matic megafauna as ‘flagship species’. In Priorities Grumbine, R.E. (1994) What is ecosystem manage- for the Conservation of Mammalian Diversity: has the ment? Conservation Biology 8: 27–38. Panda had its Day? (Eds A. Entwistle & N. Dun- Hess, G.R. & King, T.J. (2002) Plannng for wildlife in stone), pp. 53–81. Cambridge University Press, a suburbanizing landscape I: selecting focal species Cambridge. using a Delphi survey approach. Landscape and Leader-Williams, N., Harrision, J. & Green, M.J.B. Urban Planning 58: 25–40. (1990) Designing protected areas to conserve nat- Heywood, V.H. (1995) Global Biodiversity and Assess- ural resources. Science Progress 74: 189–204. ment. Cambridge University Press, New York. Lindenmayer, D.B., Manning, A.D., Smith, P.L., et al. Hoekstra, J.M., Boucher, T.M., Ricketts, T.H. & (2002) The focal-species approach and landscape Roberts, C. (2005) Confronting a biome crisis: restoration: a critique. Conservation Biology global disparities of habitat loss and protection. 16: 338–45. Ecology Letters 8: 23–9. Lunney, D., Curtin, A., Ayers, D., Cogger, H.G. & Iltis, H.H. (1988) Serendipity in the exploration of Dickman, C.R. (1996) An ecological approach to biodiversity: what good are weedy tomatoes? In identifying the endangered fauna of New South Biodiversity (Ed. E.O. Wilson), pp. 98–105. National Wales. Pacific Conservation Biology 2: 212. Academy Press, Washington, DC. Mace, G.M. & Collar, N.J. (1995) Extinction risk IUCN (2001) IUCN Red List Categories: Version 3.1. assessment for birds through quantitative criteria. Species Survival Commission, International Ibis 137: 240–46. Macdonald/Key Topics in Conservation Biology 1405122498_4_002 Final Proof page 33 6.5.2006 11:07pm

PRIORITIZING CHOICES IN CONSERVATION 33

Mace, G.M. & Lande, R. (1991) Assessing extinction Conservation Biology (Eds S. Ferson & M. Burgman), threats: toward a reevaluation of IUCN threatened pp. 291–305. Springer-Verlag, New York. species categories. Conservation Biology 5: 148–57 Possingham, H.P., Andelman, S.J., Noon, B.R., Margules, C.R. & Pressey, R.L. (2000) Systematic Trombulak, S. & Pulliam, H.R. (2001) Making smart conservation planning. Nature 405: 243–53. conservation decisions. In Research Priorities for Master, L.L. (1991) Assessing threats and setting con- Conservation Biology (Eds G. Orians & M. Soule´), servation priorities. Conservation Biology 5: 559–63. pp. 225–44. Island Press, Covelo, California. Master, L.L., Stein, B.A., Kutner, L.S. & Hammerson, Possingham, H.P., Andelman, S.J., Burgman, M.A., G. (2000) Vanishing assets: conservation status of Mendellin, R.A., Master, L.L. & Keith, D.A. (2002) US species. In Precious Heritage. The Status of Bio- Limits to the use of threatened species lists. Trends diversity in the United States (Eds B.A. Stein, L.S. in Ecology and Evolution 17: 501–7. Kutner & J.S. Adams), 93–118. Oxford University Possingham, H.P., Franklin, J., Wilson, K. & Regan, Press, New York. T.J. (2005) The roles of spatial heterogeneity and Meir, E., Andelman, S. & Possingham, H.P. (2004) ecological processes in conservation planning. In Does conservation planning matter in a dynamic Ecosystem Function in Heterogeneous Landscapes (Eds and uncertain world? Ecology Letters 7: 615–22. G.M.Lovett,C.G.Jones,M.G.Turner&K.C.Wea- Miller, B., Reading, R., Strittholt, J., et al. (1999) thers), pp. 389–406. Springer-Verlag, New York. Using focal species in the design of nature reserve Pressey, R.L. (1994) Ad hoc reservations: forward or networks. Wild Earth 8: 81–92. backward steps in developing representative re- Millsap, B.A., Gore, J.A., Runde, D.E. & Cerulean, serve systems? Conservation Biology 8: 662–8. S.I. (1990) Setting priorities for conservation of Pressey, R.L. & Taffs, K.H. (2001) Sampling of land fish and wildlife species in Florida. Wildlife types by protected areas: three measures of effect- Monographs 111: 1–57. iveness applied to western New South Wales. Bio- Mittermeier, R.A. (1986) Primate conservation prior- logical Conservation 101: 105–17. ities in the Neotropical region. In Primates: The Road Pressey, R.L., Ferrier, S., Hutchinson, C.D., Sivertsen, to Self-sustaining Populations (Ed. K. Benirschke), D.P. & Manion, G. (1995) Planning for negotiation: pp. 221–40. Springer-Verlag, New York. using an interactive geographic information system Morse, L.E. (1993) Standard and alternative taxo- to explore alternative protected area networks. In nomic data in the multi-institutional Natural Heri- Nature Conservation 4: the Role of Networks (Eds D.A. tage Data Center Network. In Designs for a Global Saunders, J.L. Craig & E.M. Mattiske), pp. 23–33. Plant Species Information System (Eds. F.A Bisby, G.F. Surrey Beatty and Sons, Sydney. Russell & R. Pankhurst), pp. 127–56. Oxford Rabinowitz, D. (1981) Seven forms of rarity. In The University Press, Oxford. Biological Aspects of Rare Plant Conservation (Ed. H. Myers, N., Mittermeier, R.A., Mittermeier, C.G., da Synge), pp. 205–17. Wiley, Chichester. Fonseca, G.A.B. & Kent, J. (2000) Biodiversity Redford, K.H., Coppolillo, P., Sanderson, E.W., et al. hotspots for conservation priorities. Nature 403: (2003) Mapping the conservation landscape. Con- 853–8. servation Biology 17: 116–31 Noss R.F., Quigley, H.B., Hornocker, M.G., Merrill, T. Reed, J.M. (1992) A system for ranking conservation & Paquet, P.C. (1996) Conservation biology and priorities for neotropical migrant birds based on conservation in the Rocky Mountains. relative susceptibility to extinction. In Ecology and Conservation Biology 10: 949–63 Conservation of Neotropical Migrant Landbirds (Eds Olson, D.M. & Dinerstein, E. (1998) The global 200: a J.M.H. Hagan III & D.W. Johnston), pp. 524–36. representation approach to conserving the Earth’s Smithsonian Institution Press, Washington, DC. most biologically valuable ecoregions. Conservation Regan, T.J., L.L. Master & G.A. Hammerson. (2004) Biology 12: 502–15. Capturing expert knowledge for threatened species Paine, R.T. (1995) A conversation on refining the assessments: a case study using Natureserve con- concept of keystone species. Conservation Biology servation status ranks. Acta Oecologica 26: 95–107. 9: 962–64. Sanderson, E.W., Redford, K.H., Chetkiewicz, Possingham, H., Ball, I. & Andelman, S. (2000) C.L.B., et al. (2002a) Planning to save a species: Mathematical methods for identifying representa- the jaguar as a model. Conservation Biology 16: tive reserve networks. In Quantitative Methods for 58–72. Macdonald/Key Topics in Conservation Biology 1405122498_4_002 Final Proof page 34 6.5.2006 11:07pm

34 G.M. MACE, H.P. POSSINGHAM AND N. LEADER-WILLIAMS

Sanderson, E.W., Redford, K.H., Vedder, A., Ward, Walpole, M.J. & Leader-Williams, N. (2002) Eco- S.E. & Coppolillo, P.B. (2002b) A conceptual tourism and flagship species in conservation. model for conservation planning based on land- Biodiversity and Conservation 11: 543–547. scape species requirements. Landscape and Urban Watson, J., Freudenberger, D. & Paull, D. (2001) An Planning 58: 41–56. assessment of the focal species approach for con- Scott, J.M., Davis, F., Csuti, B., et al. (1993) Gap serving birds in varigated landscapes in southeast- analysis: a geographic approach to protection of ern Australia. Conservation Biology 15: 1364–73. biological diversity. Wildlife Monographs 123: 1–41. Western, D. (1987) ’s elephants and rhinos: flag- Shaffer, M.L. (1981) Minimum population sizes for ships in crisis. Trends in Ecology and Evolution 2: 343–6. species conservation. BioScience, 31, 131-134 Williams, P.H., Burgess, N.D. & Rahbek, C. (2000) Simberloff, D. (1998) Flagships, umbrellas, and key- Flagship species, ecological complementarity and stones: is single-species management passe’ in the conserving the diversity of mammals and birds in landscape era. Biological Conservation 83: 247–57. sub-Saharan Africa. Animal Conservation 3: 249–60. Smith, R.J., Muir, R.D.J., Walpole, M.J., Balmford, Wilson, K.A., Westphal, M.I., Possingham, H.P. & A. & Leader-Williams, N. (2003) Governance and Elith, J. (2005) Sensitivity of conservation plan- the loss of biodiversity. Nature 426: 67–70. ning to different approaches to using predicted Stein, B.A., Kutner, L.S. & Adams, J.S. (2000) species distribution data. Biological Conservation Precious Heritage. The Status of Biodiversity in the 122: 99–112. United States. Oxford University Press, New York. Wilson, K.A., McBride, M., Bode, M. & Possingham, Stotz, D.F., Fitzpatrick, J.W., Parker, III, T.A. & H.P. (In press) Prioritising the allocation of conser- Moskovits, D.K. (1996) Neotropical Birds: Ecology and vation resources between biodiversity hotspots. Conservation. University of Chicago Press, Chicago, IL Nature.

View publication stats