Conservation opportunities across the world's anthromes

Authors


Abstract

Aim

Biologists increasingly recognize the roles of humans in ecosystems. Subsequently, many have argued that biodiversity conservation must be extended to environments that humans have shaped directly. Yet popular biogeographical frameworks such as biomes do not incorporate human land use, limiting their relevance to future conservation planning. ‘Anthromes’ map global ecological patterns created by sustained direct human interactions with ecosystems. In this paper, we set to understand how current conservation efforts are distributed across anthromes.

Location

Global.

Methods

We analysed the global distribution of IUCN protected areas and biodiversity hotspots by anthrome. We related this information to density of native plant species and density of previous ecological studies. Potential conservation opportunities in anthromes were then identified through global analysis and two case studies.

Results

Protected areas and biodiversity hotspots are not distributed equally across anthromes. Less populated anthromes contain a greater proportion of protected areas. The fewest hotspots are found within densely settled anthromes and wildlands, which occur at the two extremes of human population density. Opportunities for representative protection, prioritization, study and inclusion of native species were not congruent.

Main conclusions

Researchers and practitioners can use the anthromes framework to analyse the distribution of conservation practices at the global and regional scale. Like biomes, anthromes could also be used to set future conservation priorities. Conservation goals in areas directly shaped by humans need not be less ambitious than those in ‘natural areas’.

Introduction

An increasing number of organizations prioritize conservation projects by evaluating patterns of biodiversity and threats to biodiversity at the global scale (MA, 2005; Pereira et al., 2013). In order to identify such patterns, researchers have analysed biomes – units that map differences in vegetation type associated with variation in temperature and precipitation (Whittaker, 1962) – and ecoregions – units that incorporate data on the distribution of flora and fauna to further divide biomes into 867 distinctive biological units (Olson et al., 2001). These biogeographical frameworks have been used to identify areas of rapid habitat conversion (Hoekstra et al., 2005), biologically distinctive areas (Ricketts et al., 1999), mismatches between biological richness and endemism (Orme et al., 2005; Lamoreux et al., 2006) and patterns of ecosystem services provisioning (Naidoo et al., 2008), among other global patterns. Conservation organizations use such information to identify locations for future acquisition or management (Brooks et al., 2006). For example, Kier et al. (2005) mapped vascular plant species richness by ecoregion and compared these results with published literature on global priorities for plant conservation. Others have used biogeographical frameworks to identify ‘gaps’ in the coverage of protected areas (e.g. Rodrigues et al., 2004).

But biomes, ecoregions and related biogeographical frameworks are limited in that they reduce human influences to a single dimension of disturbance. In the past fifteen years, an increasing number of conservation biologists have argued that biodiversity conservation must be extended to habitats directly shaped by humans (Table 1). Humans dwell on or use most of Earth's land surface, and human activities affect most ecological processes (Sanderson et al., 2002; Foley et al., 2005). The diversity of approaches presented in Table 1 reflects two tenets of conservation science (Kareiva & Marvier, 2012): first, that conservation planning should be approached through multiple strategies, and second, that humans should be considered part of ecosystems.

Table 1. Examples of recent proposals asserting that future conservation actions must be extended to ecosystems directly shaped by humans and must embrace and leverage varying intensities of human activity
ProposalExample citationDescription
1. Modifying human livelihood goals to incorporate and deliver on conservation goals
AgroecologyPerrings et al. (2006)Promote integrated study of the entire food system, including ecological, economic and human dimensions
Conservation in urban areasMcKinney (2002)Develop ecologically informed public through study of urban ecology; design anthropogenic landscapes to promote biodiversity and ecosystem services
Ecosystem stewardshipChapin et al. (2010)Foster social–ecological sustainability on a rapidly changing planet
Land sharing/land sparingPhalan et al. (2011)Address biodiversity conservation alongside agronomic production goals
Reconciliation ecologyRosenzweig (2003)Redesign anthropogenic habitats so that their use is compatible with use by broad array of other species
2. Modifying conservation goals to deliver on human livelihood goals
Ecosystem-based adaptationJones et al. (2012)Use of biodiversity and ecosystem services to help people and communities adapt to the negative effects of climate change
Ecosystem servicesDaily et al. (2009)Give ‘natural capital’, which produces ecosystem services, adequate weight (monetary value) in decision-making processes
3. Redefining conservation goals given global change
Conserving the stageAnderson & Ferree (2011)Implement conservation in a manner that maximizes potential evolutionary adaptive response to climate change as opposed to trying to save all species or picking winners
Ecological connectivityKrosby et al. (2010)Soften the matrix; increase connectivity to increase probability of persistence for many organisms as climate changes
Intervention ecologyHobbs et al. (2011)Rather than attempt to restore past systems, reinstate the capacity for ecosystem functions and processes
Managed relocationRichardson et al. (2009)Save species from effects of climate change by transporting them to areas where they have not previously occurred, also termed ‘assisted colonization’ and ‘assisted migration’
Novel ecosystemsSeastedt et al. (2008)Recognize new combinations of species under new abiotic conditions (‘novel ecosystems’) and focus on desired outcomes or trajectories
RealignmentMillar & Brubaker (2006)Realign or entrain ecosystems with current and expected future conditions rather than restoring to historical pre-disturbance conditions
4. Enhancing conservation through community engagement
Citizen scienceCooper et al. (2007)Involve citizen participants directly in monitoring and management of residential lands to overcome ‘tyranny of small decisions’ to promote biodiversity
Community-based managementMcNeely (1995)Conservation planning, implementation, research, development and management must be done by local community as a whole; they must be the custodians and beneficiaries for success
Conservation bankingFox and Nino-Murcia (2005)Market that provides incentives for citizen stewardship by offering landowners opportunity to sell habitat or credits to protect private lands in perpetuity for mitigating loss of other habitat
Conservation easements and land trustsRissman et al. (2007)Protection of private lands through easements (contract that divides portions of land title between the landowner and an easement holder) and trusts (organizations that acquire and manage easements)
Human–environment interactionsMiller (2005)Design the places where people live and work to provide opportunities for meaningful interactions with the natural world
Planetary opportunitiesDeFries et al. (2012)Develop, evaluate, inform and advise society on potential pathways for sustainable development
Reserve as catalystMiller et al. (2012)Describes a model of conservation and ecosystem stewardship that is cohesive across reserves and private lands

Responding to a need for biogeographical frameworks that incorporate human processes, an increasing number of researchers and practitioners are promoting the use of integrative global-scale models of human–environment interactions (Rindfuss et al., 2004; Ellis et al., 2013), including ‘anthromes’ (anthropogenic biomes) (Ellis & Ramankutty, 2008; Václavík et al., 2013). Recently, the anthrome framework has been used to understand global ecological patterns, including the rate of landscape change over centuries (Ellis et al., 2010) and patterns of plant diversity (Ellis et al., 2012). Anthromes have also been used to highlight risks to threatened species resulting from human activity and to prioritize and direct conservation efforts in those areas. Kumara and Ramamoorthy (2011) identified anthromes in peninsular India and Sri Lanka that contained potential Travancore flying squirrel habitat. Pekin & Pijanowski (2012) combined a GIS database with distribution ranges for red-listed and non-threatened terrestrial mammal species from the IUCN with the anthromes dataset and found that the probability of endangerment increased for mammal species at the proportion of urban settlements and residential and populated croplands within their distribution ranges increased. Brum et al. (2013) found that an anthrome category (village) was a better predictor of the distribution of threatened amphibians than climate. The anthromes framework has also been suggested as a resource for conserving tropical forest biodiversity (Gardner et al., 2009), guiding management of sustainable agricultural systems (Richter & Yaalon, 2012) and identifying the relationships between multiple ecosystem services across heterogeneous land uses (Bennett et al., 2009).

Here, we use the anthrome framework to characterize current conservation efforts at the global scale. Similar analyses via biomes and ecoregions show a large bias in both the regional and global distribution of protected areas, which are concentrated primarily in areas that are at high elevations, far from human population centres or otherwise unsuitable for agriculture (Pressey, 1994; Wilson et al., 2006; Joppa & Pfaff, 2009). Recognition of this bias in distribution has resulted in efforts to develop a more representative ‘conservation portfolio’ based on biogeographical patterns (e.g. Rodrigues et al., 2004). Thus, a similar analysis under the anthrome framework will provide additional information to assure that conservation opportunities are not being missed or undervalued in coupled human–natural systems.

We begin by reviewing the human and ecological characteristics of anthromes as they pertain to biodiversity conservation. We then analyse the current global distribution of conservation efforts by anthrome, identifying underprotected, underprioritized and understudied anthromes. Through this global analysis, we identify anthromes that represent the greatest planetary opportunities (DeFries et al., 2012) for biodiversity conservation. We then present two case studies in which the anthrome framework reveals novel conservation opportunities.

What are anthromes?

Biomes are defined by differences in vegetation type associated with regional variations in climate. But humans have transformed ecosystems across more than three-quarters of the terrestrial biosphere (Sanderson et al., 2002). Anthromes (anthropogenic biomes) present an alternative view of the terrestrial biosphere by characterizing the rich diversity of global ecological land cover patterns created and sustained by human population densities and land use while also incorporating their relationships with biotic communities (Ellis & Ramankutty, 2008). For example, rather than describing the major food-, fibre- and fuel-producing regions of the world by their potential vegetation, anthromes classify these regions along gradients of population density and land use intensity to reveal ecologically important patterns (Fig. 1).

Figure 1.

Global map of contemporary anthromes (top; year 2000) and associated socio-economic and ecological characteristics, including human population density, land use and integrity of habitat and biotic communities (bottom).

Anthromes were first mapped from global data for human population, land use and remotely sensed vegetation cover using a statistical classification algorithm by Ellis & Ramankutty (2008; 21 classes). To enable long-term global changes in anthromes to be mapped, the current rule-based anthrome classification model was developed using a simplified system of 19 anthromes (Ellis et al., 2010).

Figure 1 presents a global map of contemporary anthromes, together with a conceptual diagram illustrating general patterns in human population densities and land use within and across anthromes, while also incorporating their relationships with biotic communities. Online maps and spatial data for anthromes are available for download at http://ecotope.org/anthromes/v2. Dense settlements and villages are the most populated and modified anthromes, encompassing ~70% of the global human population. Croplands, rangelands and semi-natural/forested anthromes are less densely populated but reflect significant land use change under human direction. Wildlands represent areas with little or no human activity; 85% of wildlands are located in cold and dry biomes (Ellis et al., 2010).

Within all anthromes, including densely populated anthromes, humans rarely use all available land. As a result, anthromes are generally multifunctional mosaics of heavily used lands and less intensively used lands. Growing evidence indicates that viable populations of native plants may be sustainable within anthromes, at least at local and regional scales (Ellis et al., 2012), but extending the mechanisms of conservation successes to other anthromes is necessary. Yet to date, no comparative work has explored how anthromes differ in their capacities to support biodiversity at all levels, for example species richness, functional diversity, genetic diversity (Pereira et al., 2013).

Methods

In this paper, we analyse how four ecologically important distributions map onto anthromes: protected areas (IUCN categories I-VI), prioritized areas (biodiversity hotspots; Myers et al., 2000), ecological study areas (Martin et al., 2012) and native plant richness (Ellis et al., 2012). We chose these four distributions based on their global scale, data quality and relevance to extant conservation literature. Together, they provide a snapshot of how current conservation efforts are distributed among anthromes.

We first determined the extent of protected area (IUCN categories I-VI) in each anthrome. IUCN protected areas span a wide range of protected statuses, from strict nature reserves to sustainable natural resource management areas, reflecting different intensities of human–environment interaction. Our use of ‘protected areas’ is liberal in the sense that some practitioners do not consider categories IV, V and VI to be truly protected, but is conservative in the sense that other instruments contribute to biodiversity conservation on lands not designated by IUCN status, including regulations on agricultural inputs, payments for ecosystem services, conservation easements and others.

We then determined the extent of prioritized area in each anthrome. Governments and NGOs currently employ many prioritization schemes (reviewed in Brooks et al., 2006). Here, we explore biodiversity hotspots, defined as places where large numbers of endemic vascular plant and/or vertebrate species are undergoing extensive habitat (Myers et al., 2000).

We next approximated how much research attention has been devoted to each anthrome by calculating the density of ecological field sites using a dataset developed by Martin et al. (2012). A lack of study may suggest the potential for missed biodiversity conservation opportunities. This dataset includes the spatial coordinates of the 1476 sites reported in papers published from June 2004 to June 2009 in ten top ecology journals. The dataset includes terrestrial field study sites situated world-wide, although these data are limited to studies published in English-language journals.

Finally, we calculated as the density (number km−2) of extant native plant species in an anthrome following methods described in Ellis et al. (2012). This dataset represents the first spatially explicit integrated assessment of the anthropogenic global patterns of plant species richness created by the sustained actions of human populations at regional landscape scale. In future, similar datasets may be available for other taxa.

We classified anthromes using the algorithm and data inputs of Ellis et al. (2010) from global population and land use data stratified within a geodesic Icosahedral Snyder Equal Area discrete global grid system (DGG) with a cell area of 96 km2 and intercell spacing of approximately 10.5 km (Level 12 DGG; Sahr et al., 2003). The extent of protected areas within anthromes was calculated by overlaying the 2010 WDPA area shapefile (IUCN and UNEP-WCMC 2010) on the Level 12 DGG and calculating overlay areas using a geographical information system (GIS). Hotspot areas were calculated by applying the same approach using the Level 8 DGG anthrome dataset of Ellis et al. (2012; 7792 km2 cells, intercell spacing of approximately 95 km) and the biodiversity hotspot maps of Mittermeier et al. (2003; ‘Biodiversity Hotspots’, Conservation International 2011 shapefile downloaded from: http://www.biodiversityhotspots.org/xp/Hotspots/resources/maps.xml [Accessed: 11/02/2012]). We used these results to rank anthromes by percentage protected area and percentage prioritized area.

We explored correlations between anthrome protection, prioritization, study and native species in JMP Pro 10 (SAS Institute Inc., Cary, NC, USA). We then characterized ‘conservation opportunity’ as being positively correlated with native species ranking and inversely correlated with protection, prioritization and study rankings; in other words, a lack of protection, prioritization or ecological research was taken to constitute a conservation opportunity. Finally, we summed each anthrome's four rankings to calculate a relative opportunity score for each anthrome. We then grouped anthromes into those of great conservation opportunity (1–6), moderate conservation opportunity (7–13) and lesser conservation opportunity (14–19).

Results

Protected areas are not distributed equally across anthromes at the global scale (Fig. 2). In general, less populated anthromes contain a greater proportion of protected area. For example, while 23.4% of remote woodland anthrome area is protected, only 2.3% of irrigated village anthrome area is protected.

Figure 2.

Blue bars show the total ice-free land area for each anthrome group, with the relative proportion of protected (‘Protection’) or priority hotspot (‘Prioritization’) areas shown in black. Anthromes are grouped by primary headings and subheadings.

As with protected areas, biodiversity hotspots are not distributed equally across anthromes (Fig. 2). The fewest hotspots are found within densely settled anthromes and wildlands, which occur at the two extremes of human population density. For example, hotspots encompass 1.9% of wildland anthromes, 6.7% of densely settled (urban and mixed settlement) anthromes. The greatest extent of hotspots was found in residential rainfed croplands, encompassing 38.3% of the anthrome.

The least studied anthromes are rice villages, irrigated villages and remote rangelands (Table 2). Some of the anthromes with the greatest density of native plant species are populated anthromes, including pastoral villages, mixed settlements and urban areas (Table 2).

Table 2. Anthromes ranked by opportunity for expanded representative protection, prioritization, study and density of native plant species. Those with the least opportunity are represented by ‘1’, extending to those with the greatest opportunity as represented by ‘19’. The final column (‘Opportunity’) aggregates the four rankings into one categorical scale of conservation opportunity
AnthromeProtectionPrioritizationStudyNative spp.Opportunity
Wild treeless & barren219151Moderate
Wild woodlands318112Moderate
Inhabited treeless & barren lands913166Great
Remote woodlands1171415Great
Populated woodlands41047Less
Residential woodlands79311Less
Remote rangelands516183Moderate
Populated rangelands66134Less
Residential rangelands103128Less
Remote croplands814510Moderate
Populated croplands121289Moderate
Residential rainfed croplands13575Less
Residential irrigated croplands172616Moderate
Pastoral villages154919Great
Rainfed villages16111012Great
Irrigated villages19151714Great
Rice villages1871913Great
Mixed settlements111218Less
Urban148117Moderate

Opportunities for representative protection, prioritization, study and inclusion of native species are not necessarily congruent (Table 2). For example, although a large proportion of the wild treeless and barren anthrome is protected, this anthrome category also has a low density of native plant species and is not well studied. The rice village anthrome does not contain much protected area but is prioritized under the biodiversity hotspot framework and contains a high density of native plant species. Similar to rice villages, populated and residential rangelands and residential croplands are prioritized but underprotected (Fig. 2). The ecologies of mixed settlements have yet to be explored in detail by ecologists, but these anthromes contain many native species and are relatively unprotected.

Among anthromes, proportion protected was negatively correlated with proportion prioritized under the biodiversity hotspot framework (R2 = 0.33, F1,17 = 8.34, = 0.01; Fig. 3a). In other words, an anthrome with a low proportion of protected area generally contained a high proportion of prioritized area. But proportion protected was unrelated to density of native plant species (R2 = 0.00, F1,17 = 0.00, = 0.99; Fig. 3c). Anthromes with a higher proportion prioritized under the biodiversity hotspot framework had a higher density of native plant species (R2 = 0.18, F1,17 = 3.72, = 0.07; Fig. 3c), but were not better studied by ecologists (R2 = 0.05, F1,17 = 0.95, = 0.3424; Fig. 3d).

Figure 3.

Relationships between proportion of anthrome prioritized under the biodiversity hotspot framework, proportion of anthrome protected, density of ecological study sites (sites m−2) and density of native plant species (species m−2).

Implications

At present, nine major institutional templates of global biodiversity conservation prioritization have been published (Brooks et al., 2006). Most templates prioritize high irreplaceability, but some prioritize high vulnerability and some prioritize low vulnerability – hence priority maps cover from less than one-tenth to more than one-third of the Earth's land surface. Four of these templates incorporate measures of habitat loss (Myers et al., 2000; Sanderson et al., 2002; Mittermeier et al., 2003; Hoekstra et al., 2005). Two incorporate measures of human population size (Olson & Dinerstein, 1998; Hoekstra et al., 2005). The rest do not consider human presence. But conservation biologists are increasingly prioritizing biodiversity in inhabited areas (Table 1), as are researchers in landscape planning, global change biology, agricultural science and other disciplines. Could conservation organizations use anthromes to reconsider the distribution of their projects and priorities?

Here, we demonstrate that it is possible to map multiple proxies of conservation effort by anthrome. Assuming that proportional representation by area is desirable, and weighting lack of previous prioritization, lack of previous protection, lack of previous study and density of native plant species equally, our analyses suggest that many current conservation opportunities are available within pastoral villages, rainfed villages, irrigated villages, rice villages, inhabited treeless and barren lands, and remote woodlands (Table 2). Of course, proportional representation by area across equally weighted metrics is only one of many ways to set priorities. Once an organization decides what characteristics to weight (e.g. irreplaceability, species richness) they could employ the same methods to identify anthromes of concern.

Maps of the world's anthromes enable researchers to communicate information regarding the status of global biodiversity over space and time. As a biogeographical framework, anthromes convey the conjoined fates of human populations and those of other species around the globe. Perceiving all anthromes as ecologically valuable, worthy of conservation effort and connected highlights positive actions that can be taken in landscapes previously underemphasized by conservationists.

Applications

The anthrome framework links geographically far-flung areas that, under other frameworks, would not be considered together (e.g. agroforestry in temperate and tropical populated croplands). If ecologists and conservation biologists view the world as anthromes, then new questions can be asked to study patterns in species diversity, occupancy and abundance. For example, how similar are the ecological processes and threats to biodiversity within any anthrome type across diverse environmental contexts and human cultures? Do certain anthromes present greater opportunities for ecological connectivity than others? What lessons from one anthrome that has been extensively researched (e.g. rangelands) can be applied to understudied anthromes (e.g. rice villages)?

Consider the case of the western burrowing owl (Athene cunicularia hypugaea) in the Sonoran Desert ecoregion. The Imperial Valley in California has been dramatically altered by agriculture and development and is now primarily rangeland and cropland (Fig. 4a). Yet, this region provides important habitat for the burrowing owl, harbouring approximately 70% of the state's population (DeSante et al., 2007). Research or management action would be constrained if framed under a biome or ecoregion framework, neither of which incorporates data on human land uses (Fig. 4b,c). Alternatively, a map of anthromes accounts for the dynamic and significant human land uses in this area (Fig. 4d) and draws parallels between the Imperial Valley and other anthromes encompassing mixed agricultural lands where unique conservation solutions exist (Quinn, 2013).

Figure 4.

The Imperial Valley in southern California represented by (a) current land cover, (b) biome, (c) ecoregion and (d) anthrome.

With anthromes as a backdrop, new conservation partners may be identified and new restoration techniques harnessed to extend western burrowing owl habitat into additional cropland and rangeland areas. Burrowing owls thrive on the abundant prey animals attracted to croplands and rangelands, but are limited by the lack of burrows created by fossorial mammals (Moulton et al., 2006) because intensive efforts at rodent eradication throughout croplands, rangelands and densely settled anthromes have decimated these small mammals (Marsh, 1987). In many areas, invasive annual grasses have also decreased the available habitat for fossorial mammals, and a subsequent decline in their populations has been implicated as a key factor for the decline of burrowing owls (Machicote et al., 2004). In surrounding cropland, rangeland and urban anthromes, efforts to re-establish fossorial mammals and to reduce invasive grass cover could increase habitat available to western burrowing owls. Other conservation-dependent grassland species could also benefit from efforts that are not solely focused on restoration of native habitat in wildlands.

Anthromes also provide a mechanism to identify conservation opportunities beyond known historical ranges. The California black rail (Laterallus jamaicensis coturniculus) made California's endangered species list following the conversion of San Francisco Bay-Delta Estuary tidal marshes to urban development. Thus, it was a surprise when a breeding population was documented (Aigner et al., 1995) 130 km away in Sierra Nevada rangeland (Fig. 5a), a wildland anthrome 200 years earlier (Fig. 5b). The existence of this coastal species in non-tidal marshes forced the question: how does a bird species with a perceived tidal marsh habitat requirement survive in central California? Surveys revealed that wetlands created by leaking irrigation systems provided permanent marshes necessary to sustain the endangered population and a clear instance where a wildland would not have benefited the target species. As a consequence, management recommendations and emerging policy requirements are stipulating that ranchers leave some inefficiencies (i.e. leaks) in their irrigation systems.

Figure 5.

Changes in anthrome conditions relevant to conservation of the California black rail from (a) 1800 to (b) 2000.

These brief examples highlight opportunities that would not be evident under previous biogeographical frameworks. In each case, anthromes provide a framework to simultaneously consider human activity and ecological processes and to compare conservation solutions between ecoregions (but within anthromes). Thus, the challenge is to identify the mechanisms of success in these examples and transfer them to other complex conservation situations.

Conclusion

Caro et al. (2012) have argued that acknowledgement of the Anthropocene may ‘cultivate hopelessness in those dedicated to conservation’ and excuse, or even facilitate, land use for profit. Others worry that younger generations will accept increasingly degraded environments because of ‘shifting baselines’ (Papworth et al., 2008). Alternatively, we suggest that anthrome analyses can clarify the complementary roles that conservation areas and actions can have within the biosphere. Mounting evidence is available to suggest that biodiversity conservation can be effective in both densely and sparsely settled anthromes (see references in Table 1), and biologists are increasingly questioning whether conservation goals in areas directly shaped by humans should remain less ambitious than those in ‘natural areas’. Current work seeks to address whether goals such as ecosystem services and native species preservation can be bundled (e.g. Turner et al., 2007), although, as Naidoo et al. (2008) suggest, more research is needed to evaluate the spatial concordance between areas that produce ecosystem services and those that support biodiversity. The anthromes framework could connect biodiversity conservation with extant policy instruments that focus on environment and human health outcomes in more densely populated or managed anthromes, such as payments for ecosystem services (e.g. reforestation, cropland fallowing), cost-share programmes (e.g. riparian buffers) and multistakeholder management initiatives (e.g. private lands forestry planning).

As indicated in Fig. 1, human populations tend to be concentrated in the more biodiverse regions of the Earth. Furthermore, human populations are expected to grow until at least mid-century, while per capita demand for the products of agricultural landscapes will likely increase even faster (Foley et al., 2011). Anthromes prioritized but not protected (e.g. populated and residential rangelands, residential croplands, rice villages; Fig. 2) represent areas where working outside of protected areas is most important. Ultimately, a conservation future built on maps that ignore human influences would result in missed opportunities, or a narrower field of conservation partnerships that excludes farmers, ranchers and others with an interest in conserving biodiversity in working landscapes. Rethinking the current distribution of conservation efforts by anthrome allows society to expand the horizons of biodiversity conservation. Human activity can sometimes benefit ecosystems – indeed, conservation itself is a human activity.

Acknowledgements

This project was initiated at the Ecological Society of America's second Emerging Issues conference, ‘Developing Conservation Targets under Global Change’, held in February 2012. Thanks to the conference organizers and participants, especially Nick Haddad, Andy Holdsworth, Dov Sax and Ron Swaisgood. The conference was made possible by the sponsorship of ESA, the Doris Duke Charitable Foundation, the U.S. Fish and Wildlife Service, the U.S. Geological Survey, Defenders of Wildlife and the U.S. National Park Service. This manuscript benefited from the helpful feedback of Bethany Bradley and two anonymous referees.

Biosketch

This working group comprises fourteen members from twelve research institutions across the United States and Australia. It comes out of the Ecological Society of America Emerging Issues meeting, ‘Developing Conservation Targets under Global Change’ (http://www.esa.org/pdfs/EmergingIssuesConference2012Program.pdf). The group aims to synthesize anthropogenic biomes (anthromes) and conservation data. All authors contributed to conceptual development and manuscript. L. J. Martin and J. E. Quinn led the working group and manuscript preparation.

Ancillary