Sustainable use of biodiversity - a phylogenetic perspective

Sustainable use of biodiversity requires integrated conservation of both currently-useful species and  broader phylogenetic diversity

in press, unedited accepted version of

Faith D. P. and Pollack LJ (2013) Phylogenetic diversity and the sustainable use of biodiversity. In: Applied ecology and human dimensions in biological conservation (ed. MC Lyra-Jorge and LM Verdade) Springer,

3. Phylogenetic diversity and the sustainable use of biodiversity

Daniel P. Faith
The Australian Museum
Sydney, 2010, Australia
danfaith8@yahoo.com.au
author for correspondence

Laura Jo Pollock
National Environmental Research Program
School of Botany
The University of Melbourne
Victoria 3010, Australia
laurajs@unimelb.edu.au

 

Abstract
Sustainable use of biodiversity requires the use of biodiversity in a way that does not foreclose benefits for future generations. Biodiversity option values reflect this capacity to provide future benefits that are often unanticipated. The phylogenetic diversity measure, PD, quantifies the option values represented by different sets of species. PD can be interpreted as counting-up features of species. This allows species-level ecological indices to be converted to phylogenetic indices, including PD-complementarity and PD-endemism, and integrated into systematic conservation planning. PD’s power-law relationship with species counts supports findings that initial species losses may retain high PD. This suggests that occasional loss of current-use species might not reduce overall PD. However, if species that are currently useful to society are concentrated in particular clades on the phylogeny, then their loss may imply high PD loss. Conservation of current-use species can maintain overall PD and option values. However, systematic conservation planning results suggest that conservation of phylogenetically clumped current-use species, within a given conservation budget, can produce a tipping point in which the capacity to retain high PD collapses.

3.1 Introduction
In this chapter, we will link one of the most fundamental aspects of biodiversity - the tree of life or phylogeny - to one of the most practical concerns of biodiversity conservation – the sustainable use of biodiversity. This topic contributes another perspective to our book’s overall theme on new directions for integrating applied ecology, human dimensions, and biological conservation. A precursor for this book was the 2009 Biota-FAPESP international workshop on “Applied ecology and human dimensions in biological conservation” (http://www.fapesp.br/5434). The workshop highlighted various new strategies in applied ecology, associated with emerging stronger links to human dimensions and to historical perspectives. We will touch on these themes in exploring how phylogeny helps us to understand and achieve sustainable use of biodiversity.
It is timely to consider the challenges of sustainable use of biodiversity. During 2012, the United Nations Conference on Sustainable Development (UNCSD or “Rio+20”) was held in Brazil, marking 20 years since the original conference that gave birth to the Convention on Biological Diversity. The major outcome document from the UNCSD conference refers frequently to “sustainable use of biodiversity” (UNCSD 2012). However, nearly all the references are part of a general call for “the conservation and sustainable use of biodiversity”. This invites some fresh consideration about how conservation and sustainability goals are linked.
Article 2 of the Convention on Biological Diversity (CBD; http://www.cbd.int/convention/articles/?a=cbd-02) defines “sustainable use of biodiversity” as:
“the use of biological diversity in a way and at a rate that does not lead to the long-term decline of biological diversity, thereby maintaining its potential to meet the needs and aspirations of present and future generations”.
Thus, sustainable use of biodiversity presents the challenge of providing uses to satisfy current needs while maintaining the capacity to anticipate and satisfy the needs of future generations, through other uses.
The idea that sustainable use of biodiversity requires consideration of possible future needs of society echoes the earliest justifications for biodiversity conservation, based on the idea of future prospective human uses. For example, the World Conservation Strategy (IUCN 1980) called for conservation of diversity “for present and future use”. McNeely (1988) referred to biodiversity conservation as providing a “safety net of diversity” based on its “option values” (see also Reid and Miller 1989). Option values of biodiversity are the biodiversity values that provide benefits and uses, often unanticipated, for future generations. This link means that measures of biodiversity - under the standard definition of living variation across genes, species, and ecosystems - can be interpreted as measures of option values (for review and discussion, see Faith 2012a,b; Faith 2013).
Preservation of these biodiversity option values arguably is central to any real sustainable use program. However, option values sometimes are under-appreciated in current debates about biodiversity conservation (Faith 2013). When human benefits are discussed, the term “ecosystem services” typically is used as a catch-all to cover any benefits from ecosystems that range from pristine to heavily human-modified. While ecosystem services consequently might include anything and everything, actual ecosystem services case studies typically have emphasized well-known human uses and benefits, rather than possible future uses that are currently unknown (for discussion, see Faith et al. 2010).
The focus on current essential ecosystem services also is apparent in the new Strategic Plan and 2020 Aichi targets of the Convention on Biological Diversity (www.cbd.int/doc/strategic-plan/2011-2020/Aichi-Targets-EN.pdf). These new targets provide a mixed message about the importance of benefits from biodiversity for future generations. The Mission of the Strategic Plan is “to take effective and urgent action to halt the loss of biodiversity in order to ensure that by 2020 ecosystems are resilient and continue to provide essential services.” A related Aichi target calls for maintenance of ecosystems that provide ‘essential” services. This phrasing may encourage conservation actions that focus on continued supply of those services known to be essential now, rather than worrying about future services and uses that are presently unknown and unanticipated. This same issue extends to other Aichi targets. For example, another target refers to preservation of genetic diversity, but the stated focus is on known crop species and their close relatives.
Recent characterizations of “biodiversity” reflect this popular focus on current uses. “Biodiversity” is interpreted primarily as a foundation for current uses, and biodiversity conservation sometimes seen as accomplished by ecosystem services conservation. For example, Perrings et al. (2010) suggested that ”what and how much biodiversity should be targeted for conservation depends on what services are important” (for discussion, see Faith 2011). When “important” services define the biodiversity of interest in this way, the adopted definitions and measures of biodiversity may simply re-express services in terms of their ecological basis (such as abundance and species' interactions). Traditional definitions of biodiversity recently have been expanded to include many of these aspects of species-level ecology (for discussion, see Faith 2011; 2013). For example, one ecosystem services study (Díaz et al. 2009) considered biodiversity as "the number, abundance, composition, spatial distribution, and interactions of genotypes, populations, species, functional types and traits, and landscape units in a given system". These ecological aspects may be important to the analysis of current ecosystem services, but may not help quantify option values.
Consideration of biodiversity option values sometimes is seen as less practical than strategies that link the biodiversity of interest to the ecology of ecosystem services. For example, Mace et al. (2010) argued that ‘to maintain biodiversity so as not to foreclose the options open to future generations . . .would entail a goal of no overall loss of biodiversity. . . we suggest this is unlikely to be achievable’. Others have neglected biodiversity option values, even when they do acknowledge biodiversity as something distinct from ecosystem services. In such cases, biodiversity may be characterized as primarily all about intrinsic (non-anthropocentric) values, with the human uses largely captured by the ecosystem services (for discussion, see Faith 2012a,b). In contrast, the UNCSD outcome document (UNCSD 2012), did state the importance of biodiversity values extending beyond intrinsic values:
“We reaffirm the intrinsic value of biological diversity, as well as the ecological, genetic, social, economic, scientific, educational, cultural, recreational and aesthetic values of biological diversity and its critical role in maintaining ecosystems that provide essential services…”
However, one limitation of this affirmation is that it did not explicitly highlight potential future uses or option values of biodiversity.
Some of these differences in perspective may be matters of definition (see Redford and Richter 1999). The Millennium Ecosystem Assessment (MA) (2005) has provided helpful guidelines, by distinguishing between ecosystem services and biodiversity, and by highlighting the option values of biodiversity. The MA noted that “a general lesson is that poor measurement of biodiversity reduces the capacity to discover and implement good trade-offs and synergies between biodiversity and ecosystem services. The MA also concluded that “sometimes responses to this information problem may…neglect the difficult problem of finding surrogates for global option values”.
Progress on this surrogates problem would help establish clear links between option values and sustainable use of biodiversity, and would complement both the current-uses and the intrinsic-values perspectives on biodiversity. We suggest that solutions to the “difficult” problem of finding surrogates for biodiversity option values depend on effective quantification of living variation. Such measures should include the variation among species in characters or features, because these are elements of biodiversity that might correspond to future uses and benefits (Faith 1992a; 2013). Here, we consider phylogeny, or the tree of life, as a basis for making inferences about biodiversity at this level of features of species. Our premise is that greater phylogenetic diversity, or feature diversity, implies greater option values – a greater number of potential future uses and benefits. Thus, phylogeny has particular relevance to sustainable use of biodiversity.
We consider a specific phylogenetic diversity measure PD (Faith 1992a, b) as our measure of feature diversity and option values. The PD measure not only allows us to talk about future uses but also can integrate information about current uses. Conservation of species that are currently used provides some level of conservation of the phylogenetic diversity and option values of the corresponding taxonomic group (e.g., legumes; see below). Therefore, the conservation of currently used species partially satisfies the requirements for sustainable use of biodiversity. However, a theme of this chapter is that there are advantages in integrating or balancing the conservation investments in known current-use species with conservation of broader phylogenetic diversity. We suggest that overemphasizing species that are currently valuable could reduce our capacity to preserve these broader option values – potential future uses - represented by the phylogenetic diversity within a given taxonomic group.
Our chapter is structured as follows. First, we review the phylogenetic diversity measure (PD) and the links from phylogenetic diversity to option values. Here, we highlight the need to look at gains and losses of features and PD, not just overall PD values. We show how we can replace many standard indices of species-level ecology with phylogenetic indices that count up features, not species. Second, we describe how PD calculations are relevant to the problem of conservation and sustainable use of biodiversity. Here, we describe the fundamental relationship between PD and number of species, and how this relationship changes depending on the pattern of species gains and losses across the phylogeny (i.e. phylogenetically dispersed or clumped). Third, we describe conservation planning and decision-making that integrates PD, current uses, and additional factors such as costs of conservation. We examine conservation planning scenarios using this framework. We explore the contribution of conservation of current-use species to conservation of PD, and conclude that conservation of currently used species should be complemented by direct PD conservation. We finish by returning to the general call for “the conservation and sustainable use of biodiversity” by describing how these two goals should be inter-linked through conservation planning.

3.2 PD, feature diversity, and a calculus of option values
3.2.1 Evosystem services and PD
One limitation of the ecosystem services framework is that it is very place-based in focusing on processes within ecosystems as the basis for human benefits. A complementary perspective can focus more on evolutionary processes (Faith et al. 2010; Hendry et al. 2010). Evolutionary processes, as reflected in the tree of life, generate benefits provided by characteristics or features of species. These current and future benefits for humans have been referred to as evolutionary or “evosystem services” (see Faith et al. 2010).
The phylogenetic diversity measure, “PD” (Faith 1992a, b), helps us to quantify these current and potential future benefits derived from the tree of life. The PD of a given set of species is defined as the minimum total length of all the phylogenetic branches required to connect all those species on the tree (Fig 1a). PD provides a natural way to talk about future uses and benefits provided by species because the counting-up of branch lengths links sets of species to their expected relative diversity of characters or features. PD is based on a standard model of evolutionary process that implies that shared ancestry should account for shared features (Faith 1992b). Therefore, any subset of species that has greater phylogenetic diversity, PD, will represent greater feature diversity. Because larger PD values are expected to correspond to greater feature diversity, PD values indicate option values at the level of features of species (Faith 1992a, b).
Interpretation of PD as counting-up features for different sets of species means that we also can interpret various calculations based on PD as if they are counting-up features. A family of PD measures extends conventional species-level measures and indices to the features level (Faith and Baker 2006; Faith 2008a; Nipperess et al. 2010). For example, PD-dissimilarities among localities are calculated using phylogenetic tree branches, producing measures analogous to standard Bray-Curtis and other species-level dissimilarities.
Because PD implicitly counts features among sets of species, it provides straight-forward measures of complementarity (i.e. number of additional features gained or lost) and endemism (i.e. number of features unique to a species or to an area). Complementarity and endemism values can be calculated for species or for areas. Priority setting for conservation then may focus, for example, on the PD loss if a threatened species is pruned by extinction from the phylogenetic tree (Faith 1992a, 1994). The magnitude of the PD loss from loss of any one species naturally depends on the fate of its close relatives. The loss could be large if the species were the only remaining survivor in a highly distinctive group (on the basis of PD complementarity; Fig 1b). Examples of PD-complementarity calculations are found in Forest et al. (2007) and Faith and Baker (2006). Faith (1994) provides examples of PD-endemism, including PD-endemism of amphipods for northwest Tasmania (see also Faith et al. 2004).
The PD measure is now recognised as a basis for setting conservation priorities among species or areas (Faith 1992a; Forest et al. 2007; Mace et al. 2003). Bordewich and Semple (2012) state that “phylogenetic diversity (PD) has emerged as a leading measure in quantifying the biodiversity of a collection of species.” Davies and Buckley (2011) conclude that “The loss of PD, quantified in millions of years, provides a resonant symbol of the current biodiversity crisis”.
While priorities properly focus on PD gains and losses, it is sometimes assumed that the total PD of a locality is the basis for priority setting (e.g., Isambert et al. 2011). In fact, PD complementarity and endemism are critical to such planning (Faith 1992a; Faith et al. 2004). For example, PD-complementarity is now recognised as useful for conservation planning based on molecular trees from DNA barcoding (Faith and Baker 2006). Krishnamurthy and Francis (2012) review the use of PD and DNA barcoding in conservation. In this context, Smith and Fisher (2009) document how PD calculations are important in providing robust estimates of complementarity values. We return to PD and conservation planning below.

3.2.2 Phylogenetic patterns of current uses
Some applications of PD have explicitly referred to feature diversity and option values. Examples include applications to bioprospecting, where greater PD indicates greater potential for novel discoveries (Pacharawongsakda et al. 2009; see also Saslis-Lagoudakis et al. 2011). Similarly, a study of bioprospecting of piscine venoms (Smith and Wheeler 2006) stressed the utility of phylogeny in providing predictions about unknown characteristics of species (see also Tulp and Bohlin 2002). However, the actual success of phylogeny, and the PD measure, in capturing future uses has had little investigation.
Forest et al. (2007) have provided some evidence for the utility of PD as for quantifying estimated feature diversity and option values. They examined the distribution of angiosperms plants with known human uses (classified as medicinal, food, and all other uses) on an estimated phylogenetic tree for nearly 900 genera found in the Cape hotspot of S. Africa. Their information source, the Survey of Economic Plants for Arid and Semi-Arid Lands (SEPASAL), reports on the uses of tropical and subtropical wild and semi-domesticated plants. Forest et al. (2007) labeled a given genus as ‘useful’ if it had at least one species found in the Cape and recorded in this database.
Forest et al. (2007) first asked how each use-type was distributed phylogenetically. They found that that each use-type was clumped on the tree: common ancestry often could account for taxa with the same use. This pattern suggests that phylogeny may help predict useful species, at least within any one use-type. This result corresponds to other findings. For example, Saslis-Lagoudakis et al. (2011) have found similar phylogenetic clumping for some use categories in legumes. However, Forest et al. (2007) also found that preserving species of one use-type did not do a good job of protecting species of another use-type. They found that knowledge of which plants were useful in one category would not be a good predictor of which plants were useful under another category. This suggests that protecting species with known uses generally would not be an adequate way to protect species with yet-to-be discovered uses. Forest et al. (2007) also determined that PD was the best general predictor over different use-types. Their conclusion was that current uses would not predict taxa with future uses, but that conservation of PD may effectively preserve options for the future.
These phylogenetic predictions about current and future uses highlight the role of phylogeny in capturing option values. To the extent that we are focusing on one use-type, good guesses might be made about which other species provide that use-type, based on any phylogenetic clumping of that use. Predicting which species might generally be “useful” in any of a variety of ways, is more difficult. We agree with the conclusions of Beattie et al. (2011) that “the benefits of bioprospecting have emerged from such a wide range of organisms and environments worldwide that it is not possible to predict what species or habitats will be critical to society, or industry, in the future.” As illustrated in the Forest et al. (2007) study, over a wide range of uses, it is not possible to predict which species will be useful. On the other hand, it is possible to increase our chances that a future useful species will still exist – we can do that by maximising conservation of phylogenetic diversity. Thus, we shift the goals of prediction away from specific instances to predictions about the relative amount of option value retained by different subsets of persisting species.
If conservation of PD is desirable as a way to preserve option values, then to some extent this could be achieved by retaining lots of species. In the next section, we discuss the fundamental curve linking species number to PD. We also explain why in decision-making we are interested in departures from this general curve, and how this is facilitated by integrating PD into conservation planning.

3.3 Departures from a basic PD - species relationship
3.3.1 Phylogenetically clumped or dispersed species losses
The PD of a set of species will generally increase as more species are added to the set, and it is sometimes argued that conservation priorities based on maximising species richness will also ensure conservation of phylogenetic diversity (e.g. Rodrigues and Gaston 2002). It is important therefore to consider the relationship between species number and PD, and how it varies. Faith (2008a) proposed a power law curve for the PD - species relationship (see also Faith and Williams 2006):
“The total PD represented by different-sized sets of taxa defines a “features/taxa” curve, analogous to the well-known species/area curve. Random taxon samples of different sizes from a given phylogenetic tree produced a roughly linear relationship in log–log space.”
Morlon et al. (2011) provided empirical support for this proposed power law model, based on estimated PD - species curves for four phylogenetic trees from four Mediterranean-type ecosystems. For each value of species richness (S), they calculated the PD of 100 communities obtained by randomly sampling S species from the phylogeny. This process revealed a power law PD - species relationship for all four phylogenies. This relationship is linear in log-log space (Fig. 2).
This relationship reveals some possible implications of species gains and losses on conservation of PD. One is that initial losses of species may mean only small losses in PD. At the other end of the curve, initial gains in protected species can mean large gains in PD. As the size of the protected set grows larger, the rate of gain in PD becomes progressively lower.
Those are expected patterns for the basic PD-species relationship - found when the number of species varies through random selection of species from the phylogeny. Real-world losses (and gains) will be non-random. Several studies have examined patterns of loss of phylogenetic diversity for a given number of species extinctions (reviewed in Morlon et al. 2011). The amount of actual PD loss depends in part on whether species extinctions are clumped or well-dispersed on the phylogenetic tree. For example, several studies looking at climate change impacts suggest relatively small PD losses (e.g., Yesson and Culham 2006). The climate change impacts spread out over the phylogenetic trees mean that deeper branches throughout the tree have at least one surviving descendent. Thuiller et al. (2011) similarly found small PD loss given dispersed species losses on the phylogenies for three different taxonomic groups.
In contrast, some studies have found that species extinction is concentrated on the phylogeny (“clumped”), resulting in a disproportionate loss of PD. One cause of such disproportionate loss is the occurrence of entire clades in the same threatened location. For example, this kind of clumping accounts for Huang et al.’s (2012) finding that several biodiversity hotspots in southern Asia and Amazonia are likely to lose “an unexpectedly large proportion of PD”.
Faith et al. (2010) describe phylogenetically clumped impacts as a “tipping point” problem. Successive species extinctions each may imply only a moderate loss of PD, until the last descendent species from a long branch goes extinct – and the long branch representing a large amount of PD is now lost (Fig. 3). They advocated a form of “phylogenetic risk analysis” (Faith 2008b) to guide conservation decisions that try to reduce risk of these worst case losses, or “tipping point” outcomes.
Thus, while random losses of species initially produce values near the top of the line in Fig. 2, non-random losses can produce markedly different results. Phylogenetically clumped losses may result in lower PD outcomes (Fig.2, point a), while phylogenetically well-dispersed losses could result in higher PD outcomes (Fig.2, point b). These scenarios may be relevant to the sustainable use of biodiversity. If current uses are phylogenetically clumped, as found in the study of Forest et al (2007), then loss of those species could imply a large PD loss.

3.3.2 Phylogenetically clumped or dispersed gains in species conservation
We considered a scenario above where current use of a species might lead to its loss (through some form of over-use). However, identified current uses of elements of biodiversity naturally also may act as an incentive for the conservation of those elements of biodiversity. For example, Penafiel et al. (2011) reviewed the literature on the contribution of plant and animal species to human diets and found that local food biodiversity is an important contributor of nutritious diets. They concluded that use of this variety of species in the diet has promoted the conservation of this food biodiversity.
Conservation of a set of current-use species may or may not imply the preservation of lots of PD within that taxonomic group. The PD – species curve suggests that even a small number of protected species (“gains”), selected randomly, could deliver a large gain in conserved PD. This is related to the scenario referred to above, where a small number of phylogenetically dispersed species remaining under climate change retained lots of PD. This scenario supports sustainable use – conservation of even a relatively small number of currently useful species could at the same time retain lots of PD and corresponding option values.
However, another scenario demands consideration. While phylogenetically well-dispersed gains can result in higher PD outcomes (Fig.2, point d), phylogenetically clumped gains may result in lower PD outcomes (Fig.2, point c). Considering again the Forest et al. (2007) study, the finding that current-use species are phylogenetically clumped suggests that conservation of these species may not represent much conserved PD.
We know that phylogenetically clumped impacts can imply large PD loss. It appears also that conservation of phylogenetically clumped currently-used species may not greatly help the overall conservation of PD. A solution to this problem is to somehow integrate the protection of currently used species with conservation that represents the entire phylogenetic tree for that taxonomic group. To examine this, we will explore PD, current uses, and conservation costs in systematic conservation planning.

3.4 PD in conservation planning for sustainable use
3.4.1 PD and systematic conservation planning
The “phylogenetic sustainable use problem” can be stated as follows: how do we combine conservation of overall PD (and its associated option values) with the conservation of currently valuable species? We will use PD complementarity calculations within systematic conservation planning (SCP) tools to explore this sustainable use problem.
SCP typically is recognised as a family of methods for the efficient selection of areas for the representation and persistence of elements of biodiversity (Sarkar et al. 2006). Most SCP studies are based on biodiversity measures at the species or ecosystems levels. However, phylogenetic diversity measures increasingly are considered in conservation, and methods for incorporating the PD measure into conservation planning continue to be developed. Generally, the goal is to increase the representation of PD when selecting species and/or areas for conservation and management.
The earliest PD studies (Faith 1992a; 1994) linked PD to the cornerstone of SCP, complementarity (see section 3.2.1). These early studies illustrated how PD-complementarity values could be used to efficiently select species, or areas, to add to a protected set (see also Faith and Baker 2006). The early software for this phylogenetic SCP was “PD-DIVERSITY” (Walker and Faith 1994), within the DIVERSITY package (Faith and Walker 1993). Other early phylogenetic SCP developments included the integration of costs and probabilities of extinction into PD-based priority setting (Witting and Loeschcke 1995; Weitzman 1998; Faith 2008a). Hartmann and Andre (2013) recently concluded that “using PD in a prioritisation process can typically increase biodiversity outcomes by a broad range of 10–220 %”.
While PD has been integrated into simple systematic conservation planning algorithms, there are few actual SCP applications (Rodrigues and Gaston 2002; Faith et al. 2004; Sarkar et al 2006; Strecker et al. 2011). One important development to support practical applications will be methods for taking variable costs of conservation into account. The PD-DIVERSITY software (Faith and Walker 1993; Walker and Faith 1994) implemented PD-complementarity for selecting sets of species or areas, but did not enable costs trade-offs (analyses that balance biodiversity and conservation costs).
For our trade-offs analyses for this chapter, we adopted another DIVERSITY module, TARGET (for example runs see Faith and Walker 2002). TARGET normally examines species in areas. Here, we shift the input data to features within species. An earlier example of PD systematic conservation planning using this strategy is found in Faith (2008a). The algorithm in this case builds up a list of selected species, by comparing species’ PD complementarity values to their weighted conservation costs. A species is added to the conservation set if its PD complementarity value exceeds its weighted cost. A species is deleted from the set during the course of selections if its PD complementarity value becomes less than its weighted cost. Such a deleted species initially (at an earlier stage in adding and deleting species to build up a set) may have yielded a large gain in total net benefit, but addition of other species might have reduced its biodiversity contribution, or complementarity value.
The end result of a series of additions and deletions is that the final solution, for any nominated weighting, includes a species if and only if its PD contribution exceeds its weighted cost. The final set minimizes the sum of un-represented PD and weighted cost. This selection process then is repeated for other nominated weightings, producing a trade-off curve (“efficiency frontier curve“) showing alternative solutions. We present examples of this analysis in the next section.

3.4.2 Conservation and sustainable use scenarios
Recent work has illustrated how increases in the magnitude and conservation of estimated ecosystem services can move initial high-biodiversity SCP solutions (sets of conservation priority areas) towards a tipping point in which capacity for regional biodiversity conservation collapses. This problem occurs when the areas offering ecosystem services are all much the same in their regional biodiversity contributions (Faith 2012c). This redundancy in the biodiversity of the ecosystem services areas is analogous to the clumping of evosystem services (currently used species) on a phylogeny. Do increases in the magnitude and conservation of evosystem services analogously move PD-based SCP solutions (sets of conservation priority species) towards a tipping point in which the capacity for PD conservation collapses?
Here, we present one example SCP analysis, for a simple hypothetical phylogenetic tree (Fig. 4) and assumed equal (“unit”) conservation costs for all species. We varied the assumptions about the extent and phylogenetic distribution of currently-used species. In Fig. 5a, the black efficiency frontier curve is for the case where there are no currently-used species, and the SCP analysis simply maximises PD for any nominated total cost of conservation.
We next introduced species with current uses. If the current use value of a species is assumed to imply that the cost of conservation is 0, then there is a clear gain in the net benefits obtained through SCP. For example, suppose that the first 8 members of the large clade (dots; Fig.4) have current use and are selected for conservation action at 0 cost. The green curve (Fig.5a) is the resulting efficiency frontier curve for this case where there is 0 conservation cost for the currently used species. This clearly is a desirable outcome for the sustainable use of biodiversity because a higher level of PD conservation now can be achieved for any given total cost.
However, if there is some unit cost associated with conservation of these currently used species, the trade-offs curve changes (Fig. 5a). The extent of this shift of the curve towards poor solutions depends, for any given number of current-use species, on the degree to which they are clumped on the phylogenetic tree. For example, suppose all 16 members of the large clade (Fig. 4) have current use, and are selected for conservation action, at unit cost. If there is a conservation cost for the current use species, and current use species are phylogenetically clumped, then SCP produces the red efficiency frontier curve (Fig.5a).
In Fig 5b, we summarise a range of SCP results where we maintained a constant total conservation cost (“budget”) but varied the number of currently used species. In each case, these species were phylogenetically clumped, as illustrated in Fig. 4. For this fixed budget, the plot (Fig. 5b) shows the PD conservation level identified by SCP, as the number of phylogenetically clumped, currently used, species increases. The curve shows that, for our given budget of 16 units, protecting more currently used species means much reduced overall conservation of PD. Initially, for a low number of currently used species, SCP can find high PD solutions, but as the number of currently used species increases, the capacity to represent PD drops rapidly. We conclude that conservation of currently used species, on its own, does not guarantee the retention of option values that is required for sustainable use of biodiversity. SCP analyses that integrate current uses and option values goals hold promise for achieving sustainable use, but must be monitored for the kind of tipping point we have described here.

3.5 Discussion
The PD measure reflects expected patterns of feature diversity among species and so provides a way to quantify biodiversity option values. The potential PD gains (or losses) resulting from conservation actions (or impacts) are relevant to the phylogenetic sustainable use problem. The basic PD – species curve implies that Initial species losses generally retain high PD, suggesting that occasional loss of current-use species might not reduce overall PD. For example, several species in the legume genus Pterocarpus , used in traditional medicine to treat diabetes, are now endangered (Saslis-Lagoudakis et al. 2011). However, other, closely related, species are not endangered. Therefore much of the PD of the group remains secure.
Conservation of species that have current known uses can maintain overall PD and option values; if currently used species are spread across the phylogeny, they capture more PD than those that are phylogenetically clumped.
On other occasions, current uses may be so phylogenetically clumped that losses can produce tipping points with high PD loss. This difficulty is raised by the Forest et al. (2007) study’s evidence of phylogenetically clumping of currently useful species.. Systematic conservation planning that incorporates PD potentially provides a way to overcome this problem; the relatively low PD captured by conservation of currently used species can be complemented by selected conservation of other species. Overall PD conservation then should be high. However, our systematic conservation planning results suggest an important caveat: if there is a conservation budget, conservation of lots of phylogenetically clumped current-use species can use up the budget without much conservation of PD.
Paradoxically, conserving more known-use species can reduce the capacity to conserve PD. Such undesirable sustainability tipping points may be avoided by balancing the conservation of currently-valued species and the conservation of overall phylogenetic diversity (PD). In this way, true “sustainable use” preserves not only known uses but also the sustained capacity to find other uses, in other species. We conclude that there is a need to also preserve PD as part of any program on sustainable use.
Our suggestions fit into a broader picture of the sustainable use of biodiversity. It is now well-known that management for current-use species is a major factor in the loss of biodiversity, through associated habitat loss and other factors (e.g., Lenzen et al. 2012). Thus, shared habitat (with the current-use species) is one established factor in considering biodiversity impacts and sustainable use. Here, we have shown that shared evolutionary history (with the current-use species) is another important consideration for sustainable use programs.
Brazil provides a potential good example of successful phylogenetically-based sustainable use programs. The BIOTA-FAPESP Program (www.biota.org.br/ ) has defined a goal for further development of a phylogenetic framework for exploration and assessments, in order to provide a solid basis of sustainable use of the biodiversity. The FAPESP Bioprospecta program makes effective use of phylogenies in identifying species that potentially have biologically active compounds. The FAPESP Biota program complements these efforts by also conserving overall phylogenetic diversity in the region – so supporting sustainable use by retaining options for future discovery of useful products in other species (Joly et al. 2010).
By recognizing PD as a measure of option values, and integrating conservation and use, such programs can capture the core idea (as quoted earlier from Article 2 of the CBD) that sustainable use “does not lead to the long-term decline of biological diversity, thereby maintaining its potential to meet the needs and aspirations of present and future generations.”

Acknowledgements
DF thanks the Biota Program of the São Paulo Science Foundation (FAPESP), and fellow participants at the International Workshop, in São Paulo, Brazil, on “Applied Ecology and Human Dimensions in Conservation Biology”. DF thanks participants and fellow members of the Scientific Advisory Board for discussions at the evaluation meeting of the BIOTA-FAPESP program, São Carlos, Brazil. Part of this paper was prepared for a keynote talk at the 49th Annual Meeting of the Association for Tropical Biology and Conservation (ATBC), Bonito, Mato Grosso do Sul, Brazil. DF thanks Lúcia Lohmann and the organisers for support, bioGENESIS members for discussions, and DIVERSITAS for additional funding. LP thanks the National Environmental Research Program (NERP) and the School of Botany at the University of Melbourne for support.

References
Beattie AJ, Hay M, Magnusson B, de Nys R, Smeathers J, Vincent JFV (2011) Ecology and bioprospecting. Austral Ecol 36:341-356
Bordewich M, Semple C (2012) Budgeted nature reserve selection with biodiversity feature loss and arbitrary split systems. J Math Biol 64:69-85
Davies TJ, Buckley LB (2011) Phylogenetic diversity as a window into the evolutionary and biogeographic histories of present-day richness gradients for mammals. Phil. Trans. R. Soc. B 2011 366: 2414-2425
Díaz S, Hector A, Wardle DA (2009) Biodiversity in forest carbon sequestration initiatives: not just a side benefit. Curr Opin Environ Sustainability 1:55-60
Faith DP (1992a) Conservation evaluation and phylogenetic diversity. Biol Conserv 61:1-10
Faith DP (1992b) Systematics and conservation: On predicting the feature diversity of subsets of taxa. Cladistics 8:361-373
Faith DP (1994) Phylogenetic diversity: a general framework for the prediction of feature diversity. In: Forey PL, Humphries CJ, Vane-Wright RI (Eds) Systematics and Conservation Evaluation. Clarendon Press Oxford, U.K.
Faith DP (2008a) Phylogenetic Diversity and Conservation. In: Conservation biology: evolution in action. pp. 99–115. http://books.google.com.au/books?isbn=0195306783
Faith DP (2008b) Threatened species and the preservation of phylogenetic diversity (PD): Assessments based on extinction probabilities and risk analysis. Conserv. Biol. 22: 1461-1470.
Faith DP (2010) Biodiversity transcends services. Science 330:1745-1746
Faith DP (2011) Ecosystem services and biodiversity option values. http://www.sciencemag.org/content/330/6012/1745/reply Cited 10 February 2011
Faith DP (2012a) Biodiversity and ecosystem services: similar but different. Bioscience 62: 785.
Faith DP (2012b) Common ground for biodiversity and ecosystem services: the "partial protection" challenge. F1000 Research http://f1000research.com/articles/common-ground-for-biodiversity-and-ecosystem-services-the-partial-protection-challenge/
Faith DP (2012c) Tipping points in systematic conservation planning: conservation of ecosystem services may be accompanied by a collapse in regional capacity for biodiversity conservation Abstract P2.47 in: Planet Under Pressure 2012 Conference information.
Faith DP (2013) "Biodiversity", The Stanford Encyclopedia of Philosophy, Edward N. Zalta (ed.), URL = < http://plato.stanford.edu/entries/biodiversity/ >.
Faith DP, Walker PA (1993). DIVERSITY: a software package for sampling phylogenetic and environmental diversity. Reference and user's guide. v.1.0. In. CSIRO Division of Wildlife and Ecology Canberra, Australia.
Faith DP, Walker PA (2002) The role of trade-offs in biodiversity conservation planning: linking local management, regional planning and global conservation efforts, J. Biosci. 27 (Suppl. 2) 393–407
Faith DP, Baker A (2006) Phylogenetic diversity (PD) and biodiversity conservation: some bioinformatics challenges. Evol Bioinf 2:121-128
Faith DP, Magallón S, Hendry AP, Conti E, Yahara T, Donoghue MJ (2010) Evosystem services: an evolutionary perspective on the links between biodiversity and human well-being. Curr Opin Environ Sustainability 2:66-74
Faith DP Reid CAM, Hunter J (2004) Integrating phylogenetic diversity, complementarity, and endemism for conservation assessment. Conserv Biol 18:255-261
Faith DP, Williams KJ (2006) Phylogenetic diversity and biodiversity conservation (pp. 233–235). In McGraw-Hill yearbook of science and technology. McGraw-Hill, New York, N.Y.
Forest F, Grenyer R, Rouget M, Davies TJ, Cowling RM, Faith DP, Balmford A, Manning JC, Proches S, van der Bank M, Reeves G, Hedderson TAJ, Savolainen V (2007) Preserving the evolutionary potential of floras in biodiversity hotspots. Nature 445:757-760.
Hartmann K, Andre J (2013) Should evolutionary history guide conservation?, Biodiversity and Conservation, 22:449-458.
Hendry AP, Lohmann LG, Conti E, Cracraft J, Crandall KA, Faith DP, Häuser C, Joly CA, Kogure K, Larigauderie A, Magallón S, Moritz C, Tillier S, Zardoya R, Anne-Hélène, Prieur-Richard, Walther BA, Yahara T, Donoghue MJ (2010) Evolutionary biology in biodiversity science, conservation and policy: a call to action. Evolution 64:1517-1528
Huang S, Davies TJ, Gittleman JL (2012) How global extinctions impact regional
biodiversity in mammals. Biology Letters 8: 222-225.
Isambert B, Bergsten J, Monaghan MT, Andriamizehy H, Ranarilalatiana T, Ratsimbazafy M, Andriniainimanana JR, Vogler AP (2011) Endemism and evolutionary history in conflict over Madagascar’s freshwater conservation priorities Biological Conservation 144 (2011) 1902–1909
IUCN (1980) World Conservation Strategy: living resource conservation for sustainable development, International Union for Conservation of Nature and Natural Resources (IUCN).
Joly CA, Rodrigues RR, Metzger JP, Haddad CFB, Verdade LM, Oliveira MC, Bolzani VS (2010) Biodiversity conservation research, training, and policy in São Paulo. Science 328:1358-1359.
Krishnamurthy PK, Francis RA (2012) A critical review on the utility of DNA barcoding in biodiversity conservation. Biodiversity Conservation 21:1901-1919
Lenzen M, Moran D, Kanemoto K, Foran B, Lobefaro L, Geschke A (2012) International trade drives biodiversity threats in developing nations. Nature 486:109-112
Mace GM (2003) Preserving the Tree of Life. Science 300:1707
Mace GM, Wolfgang C, Díaz S, Faith DP, Larigauderie A, Le Prestre P, Palmer M, Perrings C, Scholes RJ, Walpole M, Walther BA, Watson JEM, Mooney HA (2010) Biodiversity targets after 2010. Curr Opin Environ Sustainability 2:1-6
McNeely JA (1988) Economics and biological diversity: developing and using economic incentives to conserve biological resources. IUCN, Gland, Switzerland. 232pp.
Millennium Ecosystem Assessment (2005) Ecosystems and Human Well-being: Biodiversity Synthesis. World Resources Institute Washington DC, USA
Morlon H, Schwilk DW, Bryant JA, Marquet PA, Rebelo AG, Tauss C, Bohannan BJM, Green JL (2011) Spatial patterns in phylogenetic diversity. Ecol Lett 14:141-149
Nipperess DA, Faith DP, Barton K (2010) Resemblance in phylogenetic diversity among ecological assemblages. J Veg Sci 21: 1–12,
Pacharawongsakda E, Yokwai S, Ingsriswang S (2009) Potential natural product discovery from microbes through a diversity-guided computational framework. Appl Microbiol Biotechnol 82:579-586
Penafiel D, Lachat C, Espinel R, Van Damme P, Kolsteren P (2011) A systematic review on the contributions of edible plant and animal biodiversity to human diets. Ecohealth 8:381-399
Perrings C, Naeem S, Ahrestani F, Bunker DE, Burkill P, Canziani G, Elmqvist T, Ferrati R, Fuhrman J, Jaksic F, Kawabata Z, Kinzig A, Mace GM, Milano F, Mooney H, Prieur-Richard AH, Tschirhart J, Weisser W (2010) Conservation. Ecosystem services for 2020. Science 330:323-324
Redford KH, Richter BD (1999) Conservation of biodiversity in a world of use. Conserv Biol 13:1246-1256.
Reid WV, Miller KR (1989) Keeping Options Alive: The Scientific Basis for Conserving Biological Diversity. World Resources Institute, Washington DC.
Rodrigues ASL, Gaston K (2002) Maximising phylogenetic diversity in the selection of networks of conservation areas. Biol Conserv105: 103–111.
Sarkar S, Pressey RL, Faith DP, Margules CR, Fuller T, Stoms DM, Moffett A, Wilson KA, Williams KJ, Williams PH, Andelman S (2006) Biodiversity conservation planning tools: present status and challenges for the future. Ann Rev Environ Resour, 31:123-159
Saslis-Lagoudakis CH, Klitgaard BB, Forest F, Francis L, Savolainen V, Williamson EM, Hawkins JA (2011) The use of phylogeny to interpret cross-cultural patterns in plant use and guide medicinal plant discovery: an example from Pterocarpus (Leguminosae). PLoS ONE 6:e22275. doi: 10.1371/journal.pone.0022275
Smith MA, Fisher BL (2009) Invasions, DNA barcodes, and rapid biodiversity assessment using ants of Mauritius. Front Zool 6:31
Smith WL, Wheeler WC (2006) Venom Evolution Widespread in Fishes: A Phylogenetic Road Map for the Bioprospecting of Piscine Venoms, Journal of Heredity 97:206–217
Strecker AL, Olden JD, Whittier JB, Paukert CP (2011) Defining conservation priorities for freshwater fishes according to taxonomic, functional, and phylogenetic diversity. Ecological Applications, 21: 3002–3013
Thuiller W, Lavergne S, Roquet C, Boulangeat I, Lafourcade B, Araujo MB (2011) Consequences of climate change on the tree of life in Europe. Nature 470:531-534
Tulp M, Bohlin L (2002) Functional versus chemical diversity: is biodiversity important for drug discovery? Trends Pharmacol Sci 23:225-231
UNCSD (2012) The Future We Want. United Nations Conference on Sustainable Development Rio de Janeiro, Brazil. http://www.uncsd2012.org/content/documents/727The Future We Want 19 June 1230pm.pdf Cited 1 August 2012
Walker PA, Faith DP (1994) DIVERSITY-PD: Procedures for conservation evaluation based on phylogenetic diversity, Biodiv. Letters 2:132–139
Weitzman ML (1998) The Noah’s Ark problem. Econometrica 66:1279–1298
Witting L, Loeschcke V (1995) The optimization of biodiversity conservation. Biological Conservation 71:205–207
Yesson C, Culham A (2006) A phyloclimatic study of Cyclamen, BMC Evolutionary Biology 6:72

figure legends

Fig. 1. Hypothetical phylogenetic trees illustrating PD. a) The PD represented by the set of two species, Y and Z, as darker lines. b) Addition of species X increases the PD by the amount shown by the double line segment. This additional length needed to arrive at X is the PD-complementarity value of X.

Fig 2. A schematic diagram illustrating the power curve relationship between PD and number of species. In log-log space, this relationship is a straight line (dark line in plot). The power curve is produced by average PD values for random sets of species of a given size. Non-random sets will produce higher or lower PD values. The grey bars represent the range of possible values of PD for each number of species. Points a and c illustrate possible low-PD outcomes and points b and d illustrate possible high-PD outcomes.

Fig 3. The plot shows the PD retained as species are lost from a portion of a hypothetical phylogenetic tree having a long branch leading to three closely related species. Loss of 1 or 2 species implies only small PD loss, but loss of the 3rd species also means loss of the long ancestral branch.

Fig. 4. A hypothetical phylogenetic tree with 64 species. Species with dots are those assumed to have current uses in our analyses.

Fig 5 a) An SCP trade-offs space with vertical axis equal total PD conserved and horizontal axis equal total cost, with lower cost to right. High net benefits solutions are therefore towards the upper right. The black curve is for the case where there are no current-use species, and PD is maximised for any nominated cost. The green curve is the efficiency frontier curve for the case where there is 0 conservation cost for the current use species. The red curve is the efficiency frontier curve for the case where there is a conservation cost for the current use species, and these species are phylogenetically clumped.
b) For a fixed budget of 16 units, the plot shows the PD conservation achieved in SCP as the number of clumped current use species increases. Initially, SCP can find high PD solutions, but as the number of current use species increases the capacity to represent PD drops rapidly.

 


Dr Dan Faith , Principal Research Scientist
Last Updated: