* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download American Journal of Botan
Overexploitation wikipedia , lookup
Conservation movement wikipedia , lookup
Wildlife crossing wikipedia , lookup
Molecular ecology wikipedia , lookup
Latitudinal gradients in species diversity wikipedia , lookup
Occupancy–abundance relationship wikipedia , lookup
Soundscape ecology wikipedia , lookup
Biodiversity wikipedia , lookup
Ecological fitting wikipedia , lookup
Conservation biology wikipedia , lookup
Biogeography wikipedia , lookup
Assisted colonization wikipedia , lookup
Tropical Andes wikipedia , lookup
Landscape ecology wikipedia , lookup
Wildlife corridor wikipedia , lookup
Extinction debt wikipedia , lookup
Restoration ecology wikipedia , lookup
Source–sink dynamics wikipedia , lookup
Mission blue butterfly habitat conservation wikipedia , lookup
Biodiversity action plan wikipedia , lookup
Theoretical ecology wikipedia , lookup
Habitat destruction wikipedia , lookup
Biological Dynamics of Forest Fragments Project wikipedia , lookup
American Journal of Botany 98(3): 503–516. 2011. THE DISENTANGLED BANK: HOW LOSS OF HABITAT FRAGMENTS AND DISASSEMBLES ECOLOGICAL NETWORKS1 Andrew Gonzalez2, Bronwyn Rayfield, and Zoë Lindo Department of Biology, McGill University, 1205 Docteur Penfield, Montréal, Québec, Canada, H3A 1B1 Habitat transformation is one of the leading causes of changes in biodiversity and the breakdown of ecosystem function and services. The impacts of habitat transformation on biodiversity are complex and can be difficult to test and demonstrate. Network approaches to biodiversity science have provided a powerful set of tools and models that are beginning to present new insight into the structural and functional effects of habitat transformation on complex ecological systems. We propose a framework for studying the ways in which habitat loss and fragmentation jointly affect biodiversity by altering both habitat and ecological interaction networks. That is, the explicit study of “networks of networks” is required to understand the impacts of habitat change on biodiversity. We conduct a broad review of network methods and results, with the aim of revealing the common approaches used by landscape ecology and community ecology. We find that while a lot is known about the consequences of habitat transformation for habitat network topology and for the structure and function of simple antagonistic and mutualistic interaction networks, few studies have evaluated the consequences for large interaction networks with complex and spatially explicit architectures. Moreover, almost no studies have been focused on the continuous feedback between the spatial structure and dynamics of the habitat network and the structure and dynamics of the interaction networks inhabiting the habitat network. We conclude that theory and experiments that tackle the ecology of networks of networks are needed to provide a deeper understanding of biodiversity change in fragmented landscapes. Key words: biodiversity; extinction; graph theory; habitat fragmentation; networks; spatial ecology. One of the main scientific challenges of the 21st century is to understand the extent of biodiversity change that results from human expansion (Dirzo and Raven, 2003) and whether this change is detrimental to the functioning of the biosphere and the well-being of human society (Naeem et al., 2009). Our best data and models indicate that biodiversity is being eroded (Butchart et al., 2010). Land cover change is the primary cause of biodiversity change, and the loss of habitat area is at the core of estimates of global and regional extinction rates (e.g., Lawton and May, 1995; Pimm et al., 1995; Pimm and Raven, 2000). Other anthropogenic drivers, such as climate change and invasive species, are of growing importance, but their effects on biodiversity in the future will likely interact with habitat transformation to modify the rates of biodiversity loss (Darling and Côté, 2008). The impacts of habitat transformation on biodiversity are complex, changing not only species richness and diversity, but also the pattern of species interactions that link them in networks and the functions that species perform (Morris, 2010). Habitat transformations may follow many different trajectories involving a combination of loss, degradation, and fragmentation; hence, teasing apart the individual contributions of different types of transformations can be difficult to test and demonstrate (Ewers and Didham, 2006). Habitat fragmentation, for example, is a catch-all term for a multiscale (Keitt et al., 1997; Olff and Ritchie, 2002) process that alters habitat isolation, quality, arrangement, and connectivity (Fahrig, 2003). Difficulties in eval1 uating the effects of habitat fragmentation have led to considerable controversy surrounding the idea that habitat fragmentation per se is a significant cause of biodiversity loss (Fahrig, 2003; Yaacobi et al., 2007). Indeed, Yaacobi et al. (2007) argued that fragmentation has become a “reified doctrine” in ecology (sensu Slobodkin, 2001) because evidence for fragmentation per se as a major factor driving biodiversity loss is so weak. Recent reviews of the empirical literature (Debinski and Holt, 2000; Ewers and Didham, 2006) suggest otherwise and repeatedly find that fragmentation effects are numerous and sometimes large, but are often associated with species’ responses to other forms of environmental change that act in synergy with fragmentation. Fragmentation effects are most pronounced when the amount of habitat remaining is low (Andrén, 1994), but theses effects may be offset by spatial arrangements of habitat that promote species’ movements to create a networked system of habitat fragments (Huxel and Hastings, 1999). Network approaches (Bascompte, 2007) to biodiversity science have improved our understanding of the causes and consequences of biodiversity loss (Solé et al., 2005; Tylianakis et al., 2008). Network-based representations of ecological systems are powerful models that can describe biodiversity change (McCann, 2007) and characterize structural (e.g., Rayfield et al., in press) and functional consequences of fragmentation (Stouffer and Bascompte, 2010). Habitat transformation causes nonlinear changes in habitat network connectivity (Minor et al., 2009; Urban et al., 2009), but it also modulates the strength and timing of species interactions, that can cause cascading secondary extinctions in ecological community networks (Terborgh et al., 2001; Solé and Montoya, 2006) over extended periods of time (Helm et al., 2006; Vellend et al., 2006). Here we will review how network approaches can address key questions relating biodiversity change to the spatial structure of habitat networks and the interaction structure of ecological community networks. Steps toward a synthesis of network approaches have Manuscript received 26 October 2010; revision accepted 14 January 2011. A.G. is supported by the Canada Research Chair program and NSERC Discovery and Strategic grants. B.R. and Z.L. are supported by NSERC post-doctoral fellowships. 2 Author for correspondence (e-mail: [email protected]) doi:10.3732/ajb.1000424 American Journal of Botany 98(3): 503–516, 2011; http://www.amjbot.org/ © 2011 Botanical Society of America 503 504 American Journal of Botany [Vol. 98 been made (Holt, 2002; Polis et al., 2004; McCann et al., 2005; Amarasekare, 2008; Rooney et al., 2008). We build on these efforts by identifying promising applications of network ecology to the problem of biodiversity loss due to habitat loss and fragmentation from theoretical (Hassell et al., 1993; Bodin et al., 2006; Fortuna and Bascompte, 2006; Solé and Montoya, 2006), empirical (Komonen et al., 2000; Terborgh et al., 2001; Tylianakis et al., 2007; Bell et al., 2010), and experimental (Holyoak, 2000; Staddon et al., 2010) approaches. We close with some considerations for conservation and a framework to guide future directions for research. in fragmented landscapes that is thought to cause the extirpation of avian prey (Crooks and Soule, 1999). The fragmentation of habitat networks further affects ecosystem function as a result of changes in the flow of information, resources, and energy. For example, the loss of insect–plant pollinator interactions in fragmented landscapes can reduce the efficiency and increase the variability of this important ecosystem function (e.g., Steffan-Dewenter et al., 2006). The feedback between spatial networks and interaction networks (mutualistic or antagonistic) generates the complex dynamics typical of biodiversity change in fragmented habitats. DEFINING BIODIVERSITY AS INTERACTION NETWORKS DEFINING HABITAT FRAGMENTATION Biodiversity is the variety of living organisms, the ecological complexes in which they occur, and the ways in which they interact with each other and the environment (Groves et al., 2002). The impacts of habitat loss and fragmentation are commonly expressed through changes in species richness and the relative abundance of ecological communities. But biodiversity is also defined as variation in the number and type of interactions (e.g., predation, pollination) in which different species are engaged (Dyer et al., 2010). Some species may disproportionately facilitate biodiversity through their interactions with other species. In most ecosystems, species interactions form a vast network whose nodes and links are variable in space and time (Ings et al., 2009). For example, Petanidou et al. (2008) found that plant–pollinator interactions were highly inconstant as a result of high species turnover and changes in the degree of interaction specialization from year to year. The interaction diversity of a network can be summarized by simple measures such as link richness or more complex measures such as link asymmetry and compartmentalization. Figure 1 presents different properties of ecological network structure that have been measured in a variety of systems. These properties are highly nonrandom (McCann, 2007; Bascompte, 2009) and may be surprisingly invariant despite high species turnover (Petanidou et al., 2008). Interactions seem also to be structured in ways that may favor the robustness and resilience of the interaction networks to perturbations causing species extinction (Fortuna and Bascompte, 2006; Srinivasan et al., 2007; Allesina et al., 2009; Staniczenko et al., 2010). However, the problem of how large networks disassemble and fragment has only recently received theoretical attention (Krishna et al., 2008; Saavedra et al., 2009; Campbell et al., 2011). The fragmentation of formerly continuous habitat is expected to cause considerable node (species) extinction and transform the interaction topology of ecological networks (Fortuna and Bascompte, 2006; Solé and Montoya, 2006). This occurs, in part, because the habitat itself is converted into a network of fragments with variable spatial connectivity and quality that transforms the interaction network of species inhabiting the landscape. Only recently has the mutual influence between spatial and interaction networks been studied. A product of such a merger is the study of “networks of networks” (or “graphs-of-graphs” sensu Dale and Fortin, 2010) that emphasize the mutual influence of spatial habitat networks and species interaction networks Transformations of the interaction network also stem from the dynamics of species loss that may induce cascading effects mediated by indirect interactions. An example is mesopredator release following the extinction of large mammalian predators The conversion of vegetation for agriculture or urbanization dramatically shifts the composition of habitat in the landscape. Habitat loss is not spatially uniform and may sweep in a wave across a landscape over a period of decades (Ritters et al., 2000; Etter et al., 2006), leaving a discontinuous mosaic of remnant habitat fragments, interspersed with a mixed matrix of disturbed (e.g., intensely grazed) or regenerating vegetation. Crucially, the resulting habitat fragmentation also involves profound alterations to the physical environment that generates new habitat and great heterogeneity over multiple scales of space and time (Saunders et al., 1991; Saunders, 1998). The spectrum of organisms, from microbes to mammals, that occupies heavily fragmented landscapes experience these effects differently depending upon their size and mobility. Following Franklin et al. (2002), we define habitat fragmentation as human-induced discontinuity in the spatial distribution of resources and environmental conditions present in an area, over at least one scale, that affects the survival, reproduction (fitness), and mobility of multiple interacting species (we ignore natural causes of fragmentation such as wind and fire). The emphasis here is on the ecological response to disrupted species interactions because of changes in the connectivity of the mosaic of interspersed habitats. In natural situations, loss of connectivity co-occurs with habitat loss, which complicates the study of habitat fragmentation per se. Fahrig (2003) has called for greater rigor in the study of habitat fragmentation; few studies have clearly decomposed the contribution of changes in habitat isolation, quality, arrangement, and connectivity to biodiversity change. The effects of fragmentation, independent of habitat loss, are thought to be due in part to increased edge effects (Ewers and Didham, 2006), which can both diminish and enhance habitat quality depending upon the species (trophic) group and the spatial configuration of the habitat mosaic. For example, Fig. 2 shows the effects of spatial network structure and habitat quality on predator and prey richness in experimental moss-based landscapes (see also Chisholm et al., in press). In this experiment, habitat loss was controlled while changes in habitat connectivity and habitat patch quality were altered. The effects of spatial network form on diversity were strongest in habitat patches of poor quality within heterogeneous landscapes. Predator richness (mesostigmatid mites) only benefited in continuous landscapes, but suffered extinction when placed in fragmented landscapes (square or linear networks). Prey richness (Collembola) benefited from connectivity in all three network forms, although the effects of connectivity were less pronounced in square networks. Experimental approaches such as these coupled to theory are a powerful means of revealing the effects of changing network structure and quality on biodiversity. March 2011] Gonzalez et al.—Biodiversity networks in fragmented habitats 505 Fig. 1. Schematic representation of network structural properties that are relevant for biodiversity persistence. Examples of habitat, antagonistic, and mutualistic networks are illustrated. Network nodes represent habitat patches (black polygons, column A), species in different trophic levels (resources, R; consumers, C; and predators, P; column B), and mutualistic plant (black circles, column C) and pollinator (black squares, column C) species. Links represent potential movement routes (column A), feeding interactions (column B), and pollination interactions (column C). Link arrowheads indicate directional movements or interactions and link thickness indicates the flow rate or strength of a given link. HOW DOES HABITAT TRANSFORMATION ALTER SPATIAL HABITAT NETWORKS? indicate how to characterize the connectivity of habitat networks and quantify their network structure. In this section, we describe how the habitat networks emerge from the process of habitat loss and fragmentation. We then Modeling fragmented habitat as a network— Network approaches founded on graph theory are now increasingly applied 506 American Journal of Botany [Vol. 98 Fig. 2. The effects of network form on diversity were stronger in heterogeneous landscapes and differed among trophic groups. In this moss-based experiment, habitat loss was controlled for while changes in habitat fragmentation and network arrangement (continuous, square [looped], linear networks), and habitat quality (wet, heterogeneous, dry) were altered at the landscape level. The effects of connectivity are shown at the individual-patch level. Patches 1–4 were good quality “wet” patches in wet landscapes, patches 1–4 were poor quality “dry” patches in dry landscapes. No beneficial effects of connectivity were observed in these treatments (all values equal zero). In heterogeneous landscapes, patches 1 and 2 were good quality “wet” patches, while patches 3 and 4 were poor quality “dry” patches. (A) Predator richness (mesostigmatid mites) benefited from connectivity in heterogeneous continuous landscapes, but suffered extinction when placed in discontinuous networks of either square or linear arrangements. (B) Prey richness (collembola) benefited from connectivity in heterogeneous networks of all three forms, although the effects of connectivity were less pronounced in square networks. Values are means ± two standard errors (after Chisholm et al., in press). March 2011] Gonzalez et al.—Biodiversity networks in fragmented habitats to the problem of describing the static patterns and dynamic changes due to habitat fragmentation (reviewed by Urban et al., 2009). Habitat fragments can be represented as nodes in a spatial network connected via species dispersal to model connectivity at multiple spatial scales (Cumming et al., 2010; Dale and Fortin, 2010). These spatial habitat networks integrate several key spatial components of habitat fragmentation in their delineation and analysis: permeability of the matrix surrounding habitat fragments (McRae et al., 2008), multiple scales of dispersal among fragments (Brooks et al., 2008), and spatial configuration of habitat fragments (Andersson and Bodin, 2009). Nonlinear fragmentation in eroding habitat networks—Over time, as the total area of original habitat declines, the pattern of fragmentation in the landscape changes considerably, and metrics describing changes in the number and connectivity of remnant fragments change in a highly nonlinear manner (Forman, 1995; Trani and Giles, 1999). These observations are easily recovered by simulations based on percolation theory (Andrén, 1994; Bascompte and Solé, 1996), by which separating the role of habitat loss from habitat fragmentation per se is more easily done, and parameters that describe the change in structural fragmentation can isolate the abrupt thresholds in habitat contiguity as habitat loss progresses (Bascompte and Solé, 1996). However, critical thresholds in contiguity, predicted by random models of habitat loss, are not as common in empirical studies (Swift and Hannon, 2010), and percolation theory may be limited in its application to human-transformed landscapes (Keitt et al., 1997). Critical connectivity thresholds can be identified in models by systematically removing nodes (e.g., Bascompte et al., 2006; Bodin et al., 2006; Wilson et al., 2010) or links (e.g., Brooks, 2006; Lookingbill et al., 2010). Node-deletion sequences in habitat networks have shown a dramatic scale-dependent effect of habitat loss on landscape connectivity (Keitt et al., 1997). Habitat networks disconnect abruptly when key patches are removed that linked large habitat components (Bodin et al., 2006; Andersson and Bodin, 2009). Link-deletion sequences have demonstrated threshold effects as species’ dispersal abilities are decreased (Keitt et al., 1997). Habitat networks may suddenly fragment at relatively short distances compared to the distribution of interpatch distances (Urban and Keitt, 2001), reducing the potential for long-distance rescue within the network. In summary, critical transitions in network connectivity can occur over a narrow range of habitat loss or reduction in dispersal distances. Dispersal on habitat networks— Habitat networks have the flexibility to represent a spectrum of biological and spatial detail required to define structural, potential, or functional connectivity (Calabrese and Fagan, 2004). Field-measured dispersal rates among pairs of habitat patches should define the links in a spatial habitat network. In the absence of these data, links are defined based on knowledge of species’ movement ecology (Fagan and Calabrese, 2006; Minor and Urban, 2008). When permeability of the landscape matrix surrounding habitat patches determines species’ movements, straight-line distances between patches may no longer capture connectivity. Movement pathways may become nonlinear in which case Euclidean links and distances between patches are replaced by least-cost links and effective distances (O’Brien et al., 2006; Fall et al., 2007). Multiple links between pairs of habitat patches can be delineated (Theobald, 2006; Pinto and Keitt, 2009) and analyzed (McRae 507 et al., 2008) to account for the presence of alternative dispersal routes in fragmented landscapes. Biodiversity implies a spectrum of responses to habitat fragmentation. The many scales at which species respond to fragmentation may be addressed a priori by delineating network links based on species’ dispersal estimates (e.g., Bunn et al., 2000), or they may be identified a posteriori through a hierarchical series of networks constructed with increasing link threshold distances (Keitt et al., 1997; Brooks, 2006; Dale and Fortin, 2010). This latter approach can identify link threshold distances that correspond to abrupt changes in the connectivity pattern of the habitat network (Urban and Keitt, 2001). These threshold distances represent scales at which habitat fragments are minimally aggregated into subnets or network components (Brooks, 2006). General network properties— Network theory has identified several metrics that capture important properties of transformed habitat (Fig. 1). The six properties shown in Fig. 1 are also emphasized in network analyses of ecological interaction networks. We briefly summarize here their implications for habitat network connectivity and robustness. Link richness—Link richness provides a rudimentary indication of network connectivity. In habitat networks, absolute measures of link richness, such as the total number of links in components (Marcot and Chinn, 1982; Pascual-Hortal and Saura, 2006; Saura, 2008) or networks (Schick and Lindley, 2007; Treml et al., 2008; Saura and Torne, 2009), are common. However, relative link-richness measures have also been reported for habitat networks, most often in the form of the gamma index (Forman, 1995; Ricotta et al., 2000; Jordán, 2001) and the average node degree (Bascompte et al., 2003). The number of links in a component or network has been shown to respond to node deletion (Kininmonth et al., 2010) and addition (Brooks et al., 2008) sequences such that sparsely connected networks remain when the weaker, longer-distance links are absent. Disassortativity—Disassortative network structure results from well-connected nodes (nodes with high degrees) joining to a large number of less-well-connected nodes (nodes of low degree). Nestedness refers to the presences of disassortativity in bipartite networks (Sugihara and Ye, 2009). For example, the potential habitat of the Mexican Spotted Owl was shown to display a “hub” and “spoke” topology that indicates spatial dissasortivity (Urban and Keitt, 2001). Compartmentalization—Compartmentalization occurs when connected subnetwork groupings (i.e., compartments or modules) that are internally well connected are only loosely connected to other subnetwork groupings. Resilience is enhanced through compartmentalization because disturbances are confined within this network structure (Melián and Bascompte, 2002a). Minor and Urban (2008) specified that in their idealized reserve network highly connected nodes (“hubs”) should be spatially separated to create compartmentalization that would isolate disturbances but permit species movements across the landscape. Heterogeneous link strengths—The heterogeneity of link strengths among habitat patches has been based on movement costs of underlying land cover types (e.g., Halpin and Bunn, 2000; O’Brien et al., 2006), dispersal probability between two 508 [Vol. 98 American Journal of Botany nodes (Brooks et al., 2008; Treml et al., 2008), and the presence of multiple movement routes (McRae et al., 2008; Pinto and Keitt, 2009). Network connectivity assessments will be sensitive to least-cost definitions of links, especially in highly fragmented landscapes (Rayfield et al., 2010). Asymmetry—Habitat networks are highly asymmetric when movement between patches is directional (e.g., Treml et al., 2008) and focused into one or a few habitat fragments in the landscape. Strongly asymmetric movement patterns may arise because habitat fragments act as “ecological traps” (Schlaepfer et al., 2002), arising because of a mismatch between preference and habitat quality (Delibes et al., 2001; Robertson and Hutto, 2006) or because of a decision to select lower quality habitat (Gilroy and Sutherland, 2007). This affect may drive source– sink dynamics. In asymmetric landscapes, the persistence of populations living in sink habitats will depend strongly on their connectivity to populations living in source habitats (Urban et al., 2009). HOW DOES HABITAT LOSS PER SE CAUSE THE DISASSEMBLY OF INTERACTION NETWORKS? The effect of habitat area loss on ecological networks has been studied using scaling theory that quantifies the relationships between habitat area, species diversity, and food webs/ interaction networks. We begin with an overview of how species and link richness scale with area as a first step toward understanding how ecological networks are affected by spatial patterns. Spatial scaling—That species diversity accumulates with habitat area is a central tenet of ecology. The most common published species–area relationship (SAR) is a power function, S cA z , varied substantially between species. This interspecific variation was explained by the observation that animal-dispersed tree species were less vulnerable than wind-dispersed species. However, we are particularly concerned with how ecological networks and their properties might change with reduced habitat area. Spatial scaling theory now makes predictions for how trophic rank, food chain length (Schoener, 1989; Cohen and Newman, 1991; Holt et al., 1999; Holt, 2002; Dobson et al., 2006), and the number of trophic links (Brose et al., 2004) scale with area. Because z increases with increasing trophic level, we expect higher trophic levels to be more sensitive to habitat loss than lower trophic levels (Holt et al., 1999; Cagnolo et al., 2009). Communities should disassemble from the top-down leading to a trophic flattening of the food web in the smallest habitats (Post et al., 2000b). Extinction risk is correlated with high trophic position, larger body size, and smaller local population densities (Gilbert et al., 1998; Petchey et al., 1999). Habitat loss is known to truncate interaction networks through the loss of top predators (Terborgh et al., 2001) and parasitoids (Komonen et al., 2000), but the evidence remains equivocal (Mikkelson, 1993; Holyoak, 2000; Gonzalez and Chaneton, 2002; Rantalainen et al., 2005). A simple network property within local communities describes the power-law scaling of link richness, L, with species richness, S, as L bS u , where b and u are positive constants. Empirical studies find that u varies between 1 and 2; u = 1 suggests consumers have a finite number of resource species that is independent of consumer diversity, while u = 2 suggests that species are connected to a fixed fraction of other species and that links scale with the square of species richness. Brose et al. (2004) combined Eqs. 1 and 2 to give a simple model of the scaling of links with area: (1) where S is the number of species, A is area, c is a constant, and z is the exponent describing the rate at which species accumulate with area (recent study questions the generality of the power SAR, Guilhaumon et al., 2008). The SAR approach has been used to evaluate the effects of habitat fragmentation on biodiversity over and above the effects of habitat loss (Gonzalez, 2000; Yaacobi et al., 2007). If the original habitat area, Ao, is reduced to a smaller area, An, we do not simply expect the original number of species to decline to Sn, but rather to Sn estimated with a new higher value of z. The method involves increasing the exponent, z, of the power-relation to account for the disproportionate loss of species from small areas of habitat, whether because of small population size, reduced immigration or intense edge effects. Of course, the method is a bare bones approach and ignores a great deal of detail. Recent improvements to the quantitative framework now allow consideration of species ranges (Ney-Nifle and Mangel, 2000) and variable matrix affinity in fragmented vegetation mosaics (Pereira and Daily, 2006). Empirical estimates of biodiversity change due to habitat loss have been conducted in field studies and experiments (Spencer and Warren, 1996; Post et al., 2000a; Steffan-Dewenter et al., 2002; Steffan-Dewenter, 2003; Hoyle and Gilbert, 2004). For example, Montoya (2008) found that decreased forest cover had a strong negative effect on the occurrence of 34 common tree species. The loss of forest cover reduced tree diversity, but the response (2) L bc u Auz , (3) All else equal, larger areas should sustain communities with more links, but that relationship holds independently of S. This assumption holds for communities in which consumer and resource species co-occur spatially; however, in metacommunities or fragmented landscapes, species may be spatially separated from potential consumers or resources and thus prevent some links from occurring. Therefore, Eq. 3 cannot predict how this network property will change during broad-scale habitat loss. Brose et al. (2004) provide a unified spatial scaling relation for L that does not assume consumer-resource co-occurrence, L K θ (A)S u ck cr (4) u 1n(Sk Sr ) 1n(S ) , where K is a constant, and the subscripts k and r account for consumers and resources respectively. This relation allows for different z for each trophic level, and the term θ(A) represents the likelihood of the consumer occurring in patches with their resources, compared with random patches. This scaling relation March 2011] Gonzalez et al.—Biodiversity networks in fragmented habitats is better adapted to explain changes in link richness in fragmented landscapes, but a formal theoretical analysis of this problem has not been conducted to date. Ultimately, because space and interactions are treated implicitly, this approach cannot address how species networks are altered by habitat loss on spatial networks. Metacommunity theory has taken the step to evaluate how interaction networks disassemble under habitat loss for implicit space. Metacommunity theory— Metapopulation theory and patchdynamic models were first introduced by Levins (1969) to explain the persistence of a single species in a spatially subdivided habitat. Since then, they have been used extensively as a spatially implicit framework for studying the persistence of single species populations (Hanski and Ovaskainen, 2002), and the role of space in inducing coexistence among a large number of competitors (Levins and Culver, 1971; Hastings, 1980; May and Nowak, 1994). Despite the use of patch-dynamic models for competitive interactions, their application to trophic interactions has been more limited (Zeigler, 1977; Crowley and Murdoch, 1986). The first applications of metacommunity theory to the problem of habitat loss includes May and Nowak’s (1994) study of the effects of habitat loss on predator and prey persistence, and Tilman et al.’s (1994) elucidation of the extinction debt. These revealed that incorporating species interactions can alter our understanding of the effects of habitat loss on species persistence. The latter study predicted plant species extinctions because the destruction of habitat lowered colonization rates in fragments. Assuming that strong plant competitors are also weak colonizers, this model predicted the biased extinction of competitive dominants that can take many generations to occur. Extending this patch-dynamic model to additional trophic levels exacerbates species loss and indicates that higher trophic levels are more sensitive to habitat destruction (Wennergren et al., 1995). Melián and Bascompte (2002b) extended this metacommunity model to include simple networks in which food web modules (ranging from linear chains to apparent competition) composed of three trophic levels were studied. They found that network structure simplified from the top down in response to habitat destruction (see also Solé and Montoya, 2006), but that the degree of destruction required to eliminate the top trophic level varied between modules. Trophically mediated indirect interactions, in particular omnivory, conferred greater persistence for the top trophic level. In a recent development of the patch-dynamics framework, Pillai et al. (2010) showed that the emergence of complex food webs in metacommunities can be studied by tracking the changing patch occupancy of the various species interactions rather than patch occupancy of individual species. This general framework allows the study of arbitrarily complex networks in metacommunities undergoing habitat destruction. This framework can also incorporate mutualist interactions occurring in space (Amarasekare, 2004). Fortuna and Bascompte (2006) applied a network approach to much larger and more realistic mutualist networks and showed that the topology of mutualistic networks may make them robust to habitat loss. When compared to randomized null networks, realistic networks persisted at high levels of habitat destruction, even if they started to erode earlier than null networks. Results from patch-dynamic metacommunity models thus confirm species-area scaling theory, but also incorporate additional network complexity. 509 Metacommunity models of habitat loss have generally ignored explicit representation of space and hence spatial processes related to habitat arrangement. This is a significant gap, although exceptions exist (e.g., Hassell et al., 1993), they have focused on questions of population stability in simple host– parasitoid networks occupying landscapes undergoing habitat loss. So we are left acknowledging that much more theory is needed to understand how complex interaction networks assemble and disassemble in spatially explicit landscapes. New frameworks are now in place to study the spatial emergence of interaction network complexity (Lafferty and Dunne, 2010; Pillai et al., 2010). HOW DOES HABITAT FRAGMENTATION PER SE ALTER INTERACTION NETWORKS? The habitat area–scaling relation suggests link richness, a simple network measure, will change in landscapes as habitat area decreases. Here we summarize what is known about how link richness and other more complex interaction network structures are expected to change in fragmented landscapes. We restrict our attention to the structural properties shown in Fig. 1 and compare trophic and mutualist networks where possible. In general, we find that relatively little is known about how habitat fragmentation affects these aspects of network structure; much more research is needed before general patterns become apparent and a general theory can be tested. Link richness— The number of interactions in a food web is known to be affected by variation in the physical environment (Arim and Jaksic, 2005), but evidence for link richness responding to habitat fragmentation comes from pollinator networks. Sabatino et al. (2010) found direct effects of sierra habitat area on link richness in pollination networks persisting in transformed agricultural landscapes of the Argentine Pampas. They found that link richness increased 2-fold faster than species richness with area in pollination networks. Habitat proximity, an inverse measure of isolation, had a marginally significant influence on link but not on species richness. Disassortativity/nestedness— Two definitions of nestedness are used in the community ecology literature: (1) community nestedness and (2) interaction nestedness (i.e., disassortativity in bipartite networks). Community nestedness is the degree to which a set of communities forms different-sized subsets from the same ordered composition of species (Atmar and Patterson, 1993). Strong nestedness has been associated with communities ordered by extinction on land-bridge islands and habitat fragments of differing size (Wright et al., 2007), but this research has not examined interaction nestedness. The expectation that nestedness increases, and only increases, in communities undergoing extinction, arises from the hypothesis that species have different and predictable extinction probabilities. This hypothesis extends to changes in nestedness in interaction networks. Interaction networks are nested when species interacting with specialists are a proper subset of species interacting with generalists (Tylianakis, 2009). Specialist species, that interact with few other species, can be more sensitive to habitat fragmentation than generalists (Henle et al., 2004), and theory has shown that generalist mutualists are the least likely to suffer extinction in network simulations of habitat loss (Fortuna and Bascompte, 2006). 510 American Journal of Botany Comparmentalization/modularity— Compartmentalization has long been recognized as an important stabilizing property in food webs (Pimm and Lawton, 1980). Just as in spatial networks, compartmentalization in interaction networks occurs when internally highly connected subnetworks (i.e., compartments or modules) of species are poorly connected by none, one, or a few connector species to other subnetworks. Recent studies have detected compartments in food webs (Krause et al., 2003) and pollinator networks (Olesen et al., 2007). Bellisario et al. (2010) found three distinct compartments in detritus-based macroinvertebrate communities inhabiting a fragmented network of salt ponds. Community compartments were associated with ponds in closer spatial proximity. In this case, compartmentalization was due to species sorting among the salt ponds at different points along a salinity gradient. Habitat fragmentation can also increase compartmentalization by causing the extinction of top predators, an event that may disconnect spatially segregated interaction networks. However, no evidence for this was found within host–parasitoid webs in human-modified landscapes (Tylianakis et al., 2007). Conversely, the invasion of hypergeneralist species may connect distinct compartments and reduce compartmentalization in fragmented landscapes (Aizen et al., 2008). Surprisingly few studies have addressed the effects of compartmentalization induced by habitat fragmentation, although as we discuss below, the implications for the stability of networks dynamics are being studied (McCann et al., 2005). Interaction strength— Ecological networks may show considerable variance in the distribution of link interaction strengths (i.e., the effect of one species on the population growth rate of another). Does habitat fragmentation cause the differential extinction of species connected by strong interaction strengths? Does differential extinction of strong interactors reduce the average network interaction strength? These important questions are also not well studied despite a strong appeal by conservation science to protect highly interactive species (Soulé et al., 2003, 2005). Duffy (2003) concluded that biodiversity loss is biased toward strong interactors among animals but weak interactors among plants. Consistent with this, an extensive literature review by Tylianakis et al. (2008) suggests that interaction strengths in mutualist networks are reduced by environmental change, including habitat loss. In trophic networks, some evidence exists for increased variability in species abundances near fragment edges (Ewers and Didham, 2006). This variability may arise because interaction strengths are increased and/or because species interactions are exposed to more variable abiotic conditions (edge effects). Experiments directly measuring interaction strength in fragmented networks are now needed. Asymmetry— Symmetry in mutualist plant–animal networks occurs when specialist plants are pollinated by a few specialist pollinators and generalist plants are pollinated by different generalist pollinators. This symmetry does not appear to be the rule (Bascompte et al., 2003; Vázquez and Aizen, 2003); for example, specialist plants may interact with a several generalist pollinators (Vázquez and Simberloff, 2003). This asymmetry is essential to understand if we are to predict plant vulnerability to habitat fragmentation. Ashworth et al. (2004) argued that network asymmetry may explain why specialist and generalist plant species appear to respond similarly to habitat loss. They suggested that the biased loss of specialist pollinators in fragmented landscapes leaves both specialist and generalist plants with a similar reliance on the remaining generalist pollinators— [Vol. 98 the decline in the remaining generalist pollinators would cause similar declines in specialist and generalist plants. This perspective requires a network approach to the study of biodiversity loss in fragmented landscapes. HOW DO INTERACTION NETWORK DYNAMICS DEPEND UPON HABITAT NETWORK STRUCTURE? The effects of habitat loss on the dynamics of interaction webs have predominantly focused on colonization–extinction stability (see section How does habitat loss per se cause the disassembly of interaction networks?). This approach has revealed interaction network fragility to habitat loss and predicted the existence of delayed extinction dynamics (e.g., extinction debts), but patch-dynamic models ignore population dynamics, and additional insights are gained by studying the stability of populations embedded in interaction networks undergoing extinction (Petchey et al., 2008). The effects of habitat fragmentation are destabilizing for populations embedded in interaction networks (Hassell et al., 1993; McCann et al., 2005). These destabilizing effects are manifested as increased population variability, longer return times following disturbance, and extinction cascades that take many generations to unfold. However, spatial subdivision can stabilize highly unstable consumer–resource interactions, by introducing spatial refuges and rescue effects (Briggs and Hoopes, 2004). The effect of fragmentation per se can thus be both stabilizing and destabilizing. Rather than review this large literature here (Holt and Hoopes, 2005), we focus on how habitat loss destabilizes interaction networks by reducing the capacity of the landscape to sustain the interaction network. Spatial compression and food web instability— Mobile consumers at high trophic levels can couple distinct interaction networks in connected habitats and stabilize food web dynamics by dampening bottom up variability in resource dynamics (McCann et al., 2005; Gouhier et al., 2010). These mobile consumers persist and are themselves stabilized by averaging over asynchronous resource dynamics. A possible outcome of habitat loss is the confinement of consumer–resource networks within fragments. If habitat fragments become isolated, then the spatially extended feeding interactions of high trophic level consumers become compressed into a reduced area (McCann et al., 2005). Spatial compression generates stronger top-down trophic cascades and can cause high amplitude consumer-resource oscillations and instability. In a tritrophic network, the destabilized predator may become omnivorous by feeding on both intermediate and basal resource species (as depicted in Fig. 3). This behavioral switch to omnivory in an isolated habitat is predicted to be stabilizing, although the amount of omnivory required for stability may increase with increasing spatial compression (McCann et al., 2005). These effects point to the importance of behavioral flexibility (Abrams, 2007) as a means of dynamically altering the topology of the food web for persistence in fragmented landscapes. Habitat loss triggers instability of plant–pollinator networks— The progressive loss of habitat is known to reduce the capacity of spatial habitat networks to sustain metapopulations (Hanski and Ovaskainen, 2002). In some cases, clear thresholds of metapopulation persistence may be crossed when a critical fraction of habitat is removed. Keitt (2009) demonstrated the March 2011] Gonzalez et al.—Biodiversity networks in fragmented habitats existence of a critical habitat threshold for mutualist interaction networks between plants and pollinators. Mutualist interactions are disproportionately sensitive to habitat destruction because of inherent positive feedbacks in the form of Allee effects (see also Amarasekare 2004). Thus, scenarios involving 50–60% habitat destruction rendered the landscape incapable of sustaining the mutualist network. Spatially explicit analysis that allowed for localized interactions and movement, revealed cases of extremely slow collapse of the network; many hundreds of seasons were required to detect discernable decline even though the network was globally unstable and committed to extinction. Different patterns of habitat fragmentation (from regular to random and autocorrelated) altered persistence thresholds, with autocorrelated habitat structure favoring persistence and pollination services compared to landscapes with uncorrelated habitat distribution. WHAT ARE THE CONSERVATION IMPLICATIONS OF A NETWORK APPROACH? The interaction between habitat transformation and climate change is expected to increase the loss rate of plant (Giam et al., 2010) and animal (Jetz et al., 2007) biodiversity. Maintaining or restoring biodiversity in the face of these changes necessitates landscape-level conservation strategies that create networks of protected areas while managing the landscape matrix surrounding those reserves (Brudvig et al., 2009). A network approach is well suited to evaluate, model, and propose conservation paradigms. One of the founding paradigms in conservation biology (island biogeography theory [IBT]; MacArthur and Wilson, 1967) was that local species richness on islands (or habitat patches) is held in equilibrium not only by the resources present but also by a balance between immigration and emigration (or local extinctions). While network approaches can be applied to island models of movement among discrete, static patches of habitat, they can further incorporate new conservation paradigms pertaining to heterogeneous landscapes comprised of dynamic habitat patches embedded within a mosaic of other dynamic land cover types of varying qualities. Ultimately, these network approaches will produce future conservation paradigms about the relationship between the dynamic structural properties of networks and their functioning in terms of biodiversity maintenance and ecosystem services (May, 2006; Newman et al., 2006). The integration of current network approaches in landscape ecology and community ecology promises to provide a muchneeded acceleration in both our understanding of biodiversity loss and our ability to mitigate it. This integration will be mutually beneficial and will strengthen the arguments reviewed in this paper for the conservation of both spatial and interaction network structures (Rayfield et al., 2009; Tylianakis et al., 2010). The network tools to needed to manage and engineer landscapes for habitat connectivity are developing (McRae et al., 2008; Shah and McRae, 2008), but they do not incorporate species interaction networks. A recent application of network approaches to conservation is the design of ecosystem networks for sustainable landscapes (Opdam et al., 2006; Vos et al., 2008). Ecosystem networks are composed of mixed ecosystem types (e.g., forest, wetland, grassland etc.) linked into a spatially coherent network through the movement of organisms and resources through the landscape matrix within which it is embedded (Opdam et al., 2006). Ecosystem networks move be- 511 yond an island paradigm and combine elements of metacommunity/metaecosystem theory (Holyoak et al., 2005) with the patch-corridor and matrix landscape model of landscape ecology (Forman, 1995). It explicitly acknowledges the value of a network perspective for the management of highly fragmented landscapes for biodiversity and associated ecosystem services. Examples might include the management of a network of forest patches to sustain pollinator networks for pollination services and complex communities of natural enemies for pest control in agricultural landscapes. The empirical basis for ecosystem networks is founded upon a growing body of experimental research (Tewksbury et al., 2002; Brudvig et al., 2009; Staddon et al., 2010; Chisholm et al., in press), but to our knowledge no network has been designed and implemented with a network of species interactions in mind (network of networks, see below). Finally, the creation of ecosystem networks that are robust to climate change (Opdam and Wascher, 2004; Vos et al., 2008) will require dynamic network models that incorporate dynamic habitat and dynamic connectivity to account for the change in the species distribution, abundance, phenology, and interactions (e.g., Memmot et al., 2007) as the climate shifts in space and time (Vos et al., 2008; Williams et al., 2008). FUTURE DIRECTIONS We argued that improvement in our understanding of the effects of habitat loss and fragmentation on biodiversity will be gained if we combine the insights made using the networkbased approaches of landscape ecology (e.g., Urban et al., 2009), with those used in community ecology to capture the complex characteristics of interaction networks (Bascompte, 2009). In both fields, emphasis is placed on how the structure and strength of the links in a network affects its dynamics and functional properties. But only recently has the mutual influence between spatial and interaction networks been studied. The network properties identified and emphasized across ecological networks (Fig. 1) represent key structures hypothesized to influence network dynamics. Empirical validation of these hypotheses is required prior to usage in applications and to further develop a comprehensive theory of ecological networks. The key insight of this perspective is that there is continuous feedback between the structure and dynamics of the habitat network and the interaction networks occupying the habitat network (Fig. 3). For example, the exclusion of an apex predator (e.g., wolves) from an isolated compartment of the habitat network may release control over the herbivore (e.g., white tailed deer), initiating a trophic cascade with marked effects on lower trophic levels (vegetation biomass, species floral and faunal composition) and ecosystem services such as soil stability (Rooney and Waller, 2003; Ripple et al., 2009). This mutual influence between spatial and interaction networks is not frequently studied and is poorly understood, but its study is required if we are to clearly perceive the effects of habitat fragmentation on biodiversity. At this time, most studies have focused on the bottom-up effects of changing habitat network structure (via area or connectivity effects) on the interaction network, while the feedback involving how changes in the interaction network affect the spatial network has remained almost unexplored. Figure 3 suggests that biodiversity change in fragmented landscapes is best understood as a dynamic feedback between the topology and dynamics of habitat networks and interaction networks. We 512 [Vol. 98 American Journal of Botany Fig. 3. Visual representation of a spatial metacommunity network. A six-species interaction network (three resources, two consumers, and one predator) is spatially structured across a landscape with 10 habitat patches that are of low (white), medium (gray), or high (black) quality. Links in the habitat network represent potential movements for all species and links in the interaction networks represent feeding interactions. Shades of species’ nodes indicate the percentage of time spent in each patch, and the relative sizes of nodes indicate relative abundance. (A) The habitat network undergoes topological changes based on landscape modifications, and these spatial changes in turn constrain the topology of the global interaction network distributed across patches. Note that the illustration is for a subset of the global interaction network. Changes in the global interaction network topology influence community dynamics within habitat patches. (B) Community dynamics within a given habitat node influence the topology of the local interaction network here. Two possible outcomes of community dynamics are illustrated: omnivory and trophic cascades. Changes in the local interaction networks will in turn redefine the spatial habitat network by altering species’ movement patterns, creating a positive feedback loop (Gross and Blasius, 2008). know of no theoretical or empirical study that has studied repeated cycles around this diagram. Habitat-disturbance–succession metacommunity models (e.g., Guichard, 2005) should be easily applied to study habitat loss. These models have contributed to our understanding of self-organized spatial patterns of species distribution and abundance from local to regional scales (Pascual and Guichard, 2005). Just as in Fig. 3, these models emphasize the reciprocal interaction between pattern and process: spatial patterns of abundance feedback to influence local dynamics, which in turn influence habitat pattern. Conclusion—Two largely separate bodies of literature are currently based on the same network methods to study the impacts of habitat change on biodiversity loss. The results of this network approach clearly show that the structure and function of habitat networks and ecological interaction networks are impacted by habitat loss. Common network measures have revealed important differences in the topological robustness of antagonistic and mutualistic networks in response to habitat loss. Metacommunity models have been used to study network dynamics and in particular the disassembly and stability of interaction networks that results from habitat destruction. In general, the disassembly of ecological networks is not equivalent to their assembly, and more research is needed to understand this asymmetry (Bascompte and Stouffer, 2009). The convergence of concepts and the application of common metrics and models identified in this paper suggests that a synthesis based on the study of “networks of networks” will be achieved and will likely bring massive gains in our understanding of the complex process of biodiversity loss in human-transformed landscapes and a concomitant improvement in our capacity to mitigate the impacts of habitat loss over the coming century. LITERATURE CITED Abrams, P. A. 2007. Habitat choice in predator–prey systems: Spatial instability due to interacting adaptive movements. American Naturalist 169: 581–594. Aizen, M. A., C. L. Morales, and J. M. Morales. 2008. Invasive mutualists erode native pollination webs. PLoS Biology 6: e31. Allesina, S., A. Bodini, and M. Pascual. 2009. Functional links and robustness in food webs. Philosophical Transactions of the Royal Society, B, Biological Sciences 364: 1701–1709. March 2011] Gonzalez et al.—Biodiversity networks in fragmented habitats Amarasekare, P. 2004. Spatial dynamics of mutualistic interactions. Journal of Animal Ecology 73: 128–142. Amarasekare, P. 2008. Spatial dynamics of food webs. Annual Review of Ecology, Evolution, and Systematics 39: 479–500. Andersson, E., and O. Bodin. 2009. Practical tool for landscape planning? An empirical investigation of network based models of habitat fragmentation. Ecography 32: 123–132. Andrén, H. 1994. Effects of habitat fragmentation on birds and mammals in landscapes with different proportions of suitable habitat: A review. Oikos 71: 355–366. Arim, M., and F. M. Jaksic. 2005. Productivity and food web structure: Association between productivity and link richness among top predators. Journal of Animal Ecology 74: 31–40. Ashworth, L., R. Aguilar, L. Galetto, and M. A. Aizen. 2004. Why do pollination generalist and specialist plant species show similar reproductive susceptibility to habitat fragmentation? Journal of Ecology 92: 717–719. Atmar, W., and B. Patterson. 1993. The measure of order and disorder in the distribution of species in fragmented habitat. Oecologia 96: 373–382. Bascompte, J. 2007. Networks in ecology. Basic and Applied Ecology 8: 485–490. Bascompte, J. 2009. Disentangling the web of life. Science 325: 416–419. Bascompte, J., P. Jordano, C. J. Melián, and J. M. Olesen. 2003. The nested assembly of plant–animal mutualistic networks. Proceedings of the National Academy of Sciences, USA 100: 9383–9387. Bascompte, J., P. Jordano, and J. M. Olesen. 2006. Asymmetric coevolutionary networks facilitate biodiversity maintenance. Science 312: 431–433. Bascompte, J., and R. V. Solé. 1996. Habitat fragmentation and extinction thresholds in spatially explicit models. Journal of Animal Ecology 65: 465–473. Bascompte, J., and D. B. Stouffer. 2009. The assembly and disassembly of ecological networks. Philosophical Transactions of the Royal Society, B, Biological Sciences 364: 1781–1787. Bell, J. R., R. A. King, D. A. Bohan, and W. O. C. Symondson. 2010. Spatial co-occurrence networks predict the feeding histories of polyphagous arthropod predators at field scales. Ecography 33: 64–72. Bellisario, B., F. Cerfolli, and G. Nascetti. 2010. Spatial network structure and robustness of detritus-based communities in a patchy environment. Ecological Research 25: 813–821. Bodin, O., M. Tengo, A. Norman, J. Lundberg, and T. Elmqvist. 2006. The value of small size: Loss of forest patches and ecological thresholds in southern Madagascar. Ecological Applications 16: 440–451. Briggs, C. J., and M. F. Hoopes. 2004. Stabilizing effects in spatial parasitoid-host and predator–prey models: A review. Theoretical Population Biology 65: 299–315. Brooks, C. P. 2006. Quantifying population substructure: Extending the graph-theoretic approach. Ecology 87: 864–872. Brooks, C. P., A. Antonovics, and T. Keitt. 2008. Spatial and temporal heterogeneity explain disease dynamics in a spatially explicit network model. American Naturalist 172: 149–159. Brose, U., A. Ostling, K. Harrison, and N. D. Martinez. 2004. Unified spatial scaling of species and their trophic interactions. Nature 428: 167–171. Brudvig, L. A., E. I. Damschen, J. J. Tewksbury, N. M. Haddad, and D. J. Levey. 2009. Landscape connectivity promotes plant biodiversity spillover into non-target habitats. Proceedings of the National Academy of Sciences, USA 106: 9328–9332. Bunn, A. G., D. L. Urban, and T. H. Keitt. 2000. Landscape connectivity: A conservation application of graph theory. Journal of Environmental Management 59: 265–278. Butchart, S. H. M., M. Walpole, B. Collen, A. van Strien, J. P. W. Scharlemann, R. E. A. Almond, J. E. M. Baillie, et al. 2010. Global biodiversity: Indicators of recent declines. Science 328: 1164–1168. Cagnolo, L., G. Valladares, A. Salvo, M. Cabido, and M. Zak. 2009. Habitat fragmentation and species loss across three inter- 513 acting trophic levels: Effects of life-history and food-web traits. Conservation Biology 23: 1167–1175. Calabrese, J. M., and W. F. Fagan. 2004. A comparison-shopper’s guide to connectivity metrics. Frontiers in Ecology and the Environment 2: 529–536. Campbell, C., S. Yang, R. Albert, and K. Shea. 2011. A network model for plant–pollinator community assembly. Proceedings of the National Academy of Sciences, USA 108: 197–202. Chisholm, C., Z. Lindo, and A. Gonzalez. In press. Metacommunity diversity depends on network connectivity and arrangement in heterogeneous landscapes. Ecography [DOI: 10.1111/j.1600-0587. 2010.06588.x]. Cohen, J. E., and C. M. Newman. 1991. Community area and food-chain length: Theoretical predictions. American Naturalist 138: 1542–1554. Crooks, K. R., and M. E. Soule. 1999. Mesopredator release and avifaunal extinctions in a fragmented system. Nature 400: 563–566. Crowley, P. L., and W. W. Murdoch. 1986. Predator-mediated coexistence: An equilibrium interpretation. Journal of Theoretical Biology 80: 129–144. Cumming, G. S., O. Bodin, H. Ernstson, and T. Elmqvist. 2010. Network analysis in conservation biogeography: Challenges and opportunities. Diversity & Distributions 16: 414–425. Dale, M. R. T., and M.-J. Fortin. 2010. From graphs to spatial graphs. Annual Review of Ecology Evolution and Systematics 41: 21–38. Darling, E. S., and I. M. Côté. 2008. Quantifying the evidence for ecological synergies. Ecology Letters 11: 1278–1286. Debinski, D. M., and R. D. Holt. 2000. A survey and overview of habitat fragmentation experiments. Conservation Biology 14: 342–355. Delibes, M., P. Gaona, and P. Ferreras. 2001. Effects of an attractive sink leading into maladaptive habitat selection. American Naturalist 158: 277–285. Dirzo, R., and P. H. Raven. 2003. Global state of biodiversity and loss. Annual Review of Environment and Resources 28: 137–167. Dobson, A., D. Lodge, J. Alder, G. S. Cumming, J. Keymer, J. McGlade, H. Mooney, et al. 2006. Habitat loss, trophic collapse, and the decline of ecosystem services. Ecology 87: 1915–1924. Duffy, J. E. 2003. Biodiversity loss, trophic skew and ecosystem functioning. Ecology Letters 6: 680–687. Dyer, L. A., T. R. Walla, H. F. Greeney, J. O. Stireman III, and R. F. Hazen. 2010. Diversity of interactions: A metric for studies of biodiversity. Biotropica 42: 281–289. Etter, A., C. McAlpine, S. Phinn, D. Pullar, and H. Possingham. 2006. Characterizing a tropical deforestation wave: A dynamic spatial analysis of a deforestation hotspot in the Colombian Amazon. Global Change Biology 12: 1409–1420. Ewers, R. M., and R. K. Didham. 2006. Continuous response functions for quantifying the strength of edge effects. Journal of Applied Ecology 43: 527–536. Fagan, W. F., and J. M. Calabrese. 2006. Quantifying connectivity: Balancing metric performance with data requirements. In K. R. Crooks and M. Sanjayan [eds.], Connectivity conservation, 297–317. Cambridge University Press, Cambridge, UK. Fahrig, L. 2003. Effects of habitat fragmentation on biodiversity. Annual Review of Ecology Evolution and Systematics 34: 487–515. Fall, A., M.-J. Fortin, M. Manseau, and D. O’Brien. 2007. Spatial graphs: Principles and applications for habitat connectivity. Ecosystems 10: 448–461. Forman, R. T. T. 1995. Land mosaics: The ecology of landscapes and regions. Cambridge University Press, Cambridge, UK. Fortuna, M. A., and J. Bascompte. 2006. Habitat loss and the structure of plant–animal mutualistic networks. Ecology Letters 9: 281–286. Franklin, A. B., B. R. Noon, and T. L. George. 2002. What is habitat fragmentation? Studies in Avian Biology 25: 20–29. Giam, X., C. J. A. Bradshaw, H. T. W. Tan, and N. S. Sodhi. 2010. Future habitat loss and the conservation of plant biodiversity. Biological Conservation 143: 1594–1602. Gilbert, F., A. Gonzalez, and I. Evans-Freke. 1998. Corridors maintain species richness in the fragmented landscapes of a microecosystem. 514 American Journal of Botany Proceedings of the Royal Society of London, B, Biological Sciences 265: 577–582. Gilroy, J. J., and W. J. Sutherland. 2007. Beyond ecological traps: Perceptual errors and undervalued resources. Trends in Ecology & Evolution 22: 351–356. Gonzalez, A. 2000. Community relaxation in fragmented landscapes: The relation between species richness, area and age. Ecology Letters 3: 441–448. Gonzalez, A., and E. J. Chaneton. 2002. Heterotroph species extinction, abundance and biomass dynamics in an experimentally fragmented microecosystem. Journal of Animal Ecology 71: 594–602. Gouhier, T. C., F. Guichard, and A. Gonzalez. 2010. Synchrony and stability of food webs in metacommunities. American Naturalist 175: E16–E34. Gross, T., and B. Blasius. 2008. Adaptive coevolutionary networks: A review. Journal of the Royal Society, Interface 5: 259–271. Groves, C. R., D. B. Jensen, L. L. Valutis, K. H. Redford, M. L. Shaffer, J. M. Scott, J. V. Baumgartner, et al. 2002. Planning for biodiversity conservation: Putting conservation science into practice. BioScience 52: 499–512. Guichard, F. 2005. Interaction strength and extinction risk in a metacommunity. Proceedings of the Royal Society of London, B, Biological Sciences 272: 1571–1576. Guilhaumon, F., O. Gimenez, K. J. Gaston, and D. Mouillot. 2008. Taxonomic and regional uncertainty in species–area relationships and the identification of richness hotspots. Proceedings of the National Academy of Sciences, USA 105: 15458–15463. Halpin, P. N., and A. G. Bunn. 2000. Using GIS to compute a least-cost distance matrix: A comparison of terrestrial and marine ecological applications. In Proceedings of the 20th annual ESRI User Conference, 1–19, 2000, San Diego, California, USA. Esri, Redlands, California, USA Hanski, I., and O. Ovaskainen. 2002. Extinction debt at extinction threshold. Conservation Biology 16: 666–673. Hassell, M. P., H. N. C. Godfray, and H. N. Coinins. 1993. Effects of global change on the dynamics of insect host–parasitoid interactions. In P. Kareiva and J. Kingsolver [eds.], Biotic interactions and global change, 402–423. Sinauer, Sunderland, Massachusetts, USA. Hastings, A. 1980. Disturbance, coexistence, history, and competition for space. Theoretical Population Biology 18: 363–373. Helm, A., I. Hanski, and M. Pärtel. 2006. Slow response of plant species richness to habitat loss and fragmentation. Ecology Letters 9: 72–77. Henle, K., K. F. Davies, M. Kleyer, C. Margules, and J. Settele. 2004. Predictors of species sensitivity to fragmentation. Biodiversity and Conservation 13: 207–251. Holt, R. D. 2002. Food webs in space: On the interplay of dynamic instability and spatial processes. Ecological Research 17: 261–273. Holt, R. D., and M. F. Hoopes. 2005. Food web dynamics in a metacommunity context: Modules and beyond. In M. Holyoak and M. A. Leibold [eds.], Metacommunities: Spatial dynamics and ecological communities, 68-93. University of Chicago Press, Chicago, Illinois, USA. Holt, R. D., J. H. Lawton, G. A. Polis, and N. D. Martinez. 1999. Trophic rank and the species–area relationship. Ecology 80: 1495–1504. Holyoak, M. 2000. Habitat patch arrangement and metapopulation persistence of predators and prey. American Naturalist 156: 378–389. Holyoak, M., M. A. Leibold, and R. D. Holt. 2005. Metacommunities: Spatial dynamics and ecological communities. University of Chicago Press, Chicago, Illinois, USA. Hoyle, M., and F. Gilbert. 2004. Species richness of moss landscapes unaffected by short-term fragmentation. Oikos 105: 359–367. Huxel, G. R., and A. Hastings. 1999. Habitat loss, fragmentation, and restoration. Restoration Ecology 7: 309–315. Ings, T. C., J. M. Montoya, J. Bascompte, N. Bluthgen, L. Brown, C. F. Dormann, F. Edwards, et al. 2009. Ecological networks— Beyond food webs. Journal of Animal Ecology 78: 253–269. [Vol. 98 Jetz, W., D. S. Wilcove, and A. P. Dobson. 2007. Predicted impacts of climate change and land-use change on the global biodiversity of birds. PLoS Biology 5: e157. Jordán, F. 2001. Adding function to structure—Comments on Palmarola landscape connectivity. Community Ecology 2: 133–135. Keitt, T. H. 2009. Habitat conversion, extinction thresholds, and pollination services in agroecosystems. Ecological Applications 19: 1561–1573. Keitt, T. H., D. L. Urban, and B. T. Milne. 1997. Detecting critical scales in fragmented landscapes [online]. Ecology and Society 1: http://www.consecol.org/vol1/iss1/art4/. Kininmonth, S. J., G. De’ath, and H. P. Possingham. 2010. Graph theoretic topology of the Great but small Barrier Reef world. Theoretical Ecology 3: 75–88. Komonen, A., R. Penttilä, M. Lindgren, and I. Hanski. 2000. Forest fragmentation truncates a food chain based on an old-growth forest bracket fungus. Oikos 90: 119–126. Krause, A. E., K. A. Frank, D. M. Mason, R. E. Ulanowicz, and W. W. Taylor. 2003. Compartments revealed in food-web structure. Nature 426: 282–285. Krishna, A., P. R. Guimaraes, P. Jordano, and J. Bascompte. 2008. A neutral-niche theory of nestedness in mutualistic networks. Oikos 117: 1609–1618. Lafferty, K. D., and J. A. Dunne. 2010. Stochastic ecological network occupancy (SENO) models: A new tool for modeling ecological networks across spatial scales. Theoretical Ecology 3: 123–135. Lawton, J. H., and R. M. May. 1995. Extinction rates. Oxford University Press, Oxford, UK. Levins, R. 1969. Some demographic and genetic consequences of environmental heterogeneity for biological control. Bulletin of the Entomological Society of America 15: 237–240. Levins, R., and D. Culver. 1971. Regional coexistence of species and competition between rare species. Proceedings of the National Academy of Sciences, USA 68: 1246–1248. Lookingbill, T. R., R. H. Gardner, J. R. Ferrari, and C. E. Keller. 2010. Combining a dispersal model with network theory to assess habitat connectivity. Ecological Applications 20: 427–441. MacArthur, R. H., and E. O. Wilson. 1967. The theory of island biogeography. Princeton University Press, Princeton, New Jersey, USA. Marcot, B. G., and P. Z. Chinn. 1982. Use of graph theory measures for assessing diversity of wildlife habitat. In R. Lamberson [ed.], Mathematical models of renewable resources, Proceedings of the First Pacific Coast Conference on Mathematical Models of Renewable Resources. Humboldt State University, Arcata, California, USA. May, R. M. 2006. Network structure and the biology of populations. Trends in Ecology & Evolution 21: 394–399. May, R. M., and M. A. Nowak. 1994. Superinfection, metapopulation dynamics and the evolution of diversity. Journal of Theoretical Biology 170: 95–114. McCann, K. S. 2007. Protecting biostructure. Nature 446: 29. McCann, K. S., J. B. Rasmussen, and J. Umbanhowar. 2005. The dynamics of spatially coupled food webs. Ecology Letters 8: 513–523. McRae, B. H., B. G. Dickson, T. H. Keitt, and V. B. Shah. 2008. Using circuit theory to model connectivity in ecology, evolution, and conservation. Ecology 89: 2712–2724. Melián, C. J., and J. Bascompte. 2002a. Complex networks: Two ways to be robust? Ecology Letters 5: 705–708. Melián, C. J., and J. Bascompte. 2002b. Food web structure and habitat loss. Ecology Letters 5: 37–46. Memmott, J., P. G. Craze, N. M. Waser, and M. V. Price. 2007. Global warming and the disruption of plant–pollinator interactions. Ecology Letters 10: 710–717. Mikkelson, G. M. 1993. How do food webs fall apart—A study of changes in trophic structure during relaxation on habitat fragments. Oikos 67: 539–547. Minor, E. S., S. M. Tessel, K. A. M. Engelhardt, and T. R. Lookingbill. 2009. The role of landscape connectivity in assembling exotic plant communities: A network analysis. Ecology 90: 1802–1809. March 2011] Gonzalez et al.—Biodiversity networks in fragmented habitats Minor, E. S., and D. L. Urban. 2008. A graph-theory frarmework for evaluating landscape connectivity and conservation planning. Conservation Biology 22: 297–307. Montoya, D. 2008. Habitat loss, dispersal, and the probability of extinction of tree species. Communicative & Integrative Biology 1: 146–147. Morris, R. J. 2010. Anthropogenic impacts on tropical forest biodiversity: A network structure and ecosystem functioning perspective. Philosophical Transactions of the Royal Society, B, Biological Sciences 365: 3709–3718. Naeem, S., D. E. Bunker, A. Hector, M. Loreau, and C. Perrings. 2009. Biodiversity, ecosystem functioning, and human wellbeing: An ecological and economic perspective. Oxford University Press, Oxford, UK. Newman, M., A.-L. Barabasi, and D. Watts. 2006. The structure and dynamics of networks. Princeton University Press, Princeton, New Jersey, USA. Ney-Nifle, M., and M. Mangel. 2000. Habitat loss and changes in the species–area relationship. Conservation Biology 14: 893–898. O’Brien, D., M. Manseau, A. Fall, and M.-J. Fortin. 2006. Testing the importance of spatial configuration of winter habitat for woodland caribou: An application of graph theory. Biological Conservation 130: 70–83. Olesen, J. M., J. Bascompte, Y. L. Dupont, and P. Jordano. 2007. The modularity of pollination networks. Proceedings of the National Academy of Sciences, USA 104: 19891–19896. Olff, H., and M. E. Ritchie. 2002. Fragmented nature: Consequences for biodiversity. Landscape and Urban Planning 58: 83–92. Opdam, P., E. Steingröver, and S. van Rooij. 2006. Ecological networks: A spatial concept for multi-actor planning of sustainable landscapes. Landscape and Urban Planning 75: 322–332. Opdam, P., and D. Wascher. 2004. Climate change meets habitat fragmentation: Linking landscape and biogeographical scale level in research and conservation. Biological Conservation 117: 285–297. Pascual, M., and F. Guichard. 2005. Criticality and disturbance in spatial ecological systems. Trends in Ecology & Evolution 20: 88–95. Pascual-Hortal, L., and S. Saura. 2006. Comparison and development of new graph-based landscape connectivity indices: Towards the priorization of habitat patches and corridors for conservation. Landscape Ecology 21: 959–967. Pereira, H. M., and G. C. Daily. 2006. Modeling biodiversity dynamics in countryside landscapes. Ecology 87: 1877–1885. Petanidou, T., A. S. Kallimanis, J. Tzanopoulos, S. P. Sgardelis, and J. D. Pantis. 2008. Long-term observation of a pollination network: Fluctuation in species and interactions, relative invariance of network structure and implications for estimates of specialization. Ecology Letters 11: 564–575. Petchey, O. L., A. P. Beckerman, J. O. Riede, and P. H. Warren. 2008. Size, foraging, and food web structure. Proceedings of the National Academy of Sciences, USA 105: 4191–4196. Petchey, O. L., P. T. McPhearson, T. M. Casey, and P. J. Morin. 1999. Environmental warming alters food-web structure and ecosystem function. Nature 402: 69–72. Pillai, P., M. Loreau, and A. Gonzalez. 2010. A patch-dynamic framework for food web metacommunities. Theoretical Ecology 3: 223–237. Pimm, S. L., and J. H. Lawton. 1980. Are food webs divided into compartments? Journal of Animal Ecology 49: 879–898. Pimm, S. L., and P. H. Raven. 2000. Biodiversity: Extinction by numbers. Nature 403: 843–845. Pimm, S. L., G. J. Russell, J. L. Gittleman, and T. M. Brooks. 1995. The future of biodiversity. Science 269: 347–350. Pinto, N., and T. Keitt. 2009. Beyond the least-cost path: Evaluating corridor redundancy using a graph-theoretic approach. Landscape Ecology 24: 253–266. Polis, G. A., M. E. Power, and G. R. Huxel. 2004. Food webs at the landscape level. University of Chicago Press, Chicago, Illinois, USA. Post, D. M., M. E. Conners, and D. S. Goldberg. 2000a. Prey preference by a top predator and the stability of linked food chains. Ecology 81: 8–14. 515 Post, D. M., M. L. Pace, and N. G. Hairston. 2000b. Ecosystem size determines food-chain length in lakes. Nature 405: 1047–1049. Rantalainen, M. L., H. Fritze, J. Haimi, T. Pennanen, and H. Setälä. 2005. Species richness and food web structure of soil decomposer community as affected by the size of habitat fragment and habitat corridors. Global Change Biology 11: 1614–1627. Rayfield, B., M.-J. Fortin, and A. Fall. 2010. The sensitivity of leastcost habitat graphs to relative cost surface values. Landscape Ecology 25: 519–532. Rayfield, B., M.-J. Fortin, and A. Fall. In press. Connectivity for conservation: A framework to classify network measures. Ecology doi: 10.1890/09-2190.1 Rayfield, B., A. Moilanen, and M.-J. Fortin. 2009. Incorporating consumer–resource spatial interactions in reserve design. Ecological Modelling 220: 725–733. Ricotta, C., A. Stanisci, G. Avena, and C. Blasi. 2000. Quantifying the network connectivity of landscape mosaics: A graph-theoretical approach. Community Ecology 1: 89–94. Ripple, W. J., T. P. Rooney, and R. L. Beschta. 2009. Large predators, deer, and trophic cascades in boreal and temperate ecosystems. In J. Terborgh and J. Estes [eds.], Trophic cascades: Predators, prey, and the changing dynamics of nature, 141–161. Island Press, Washington, D.C., USA. Ritters, K., J. Wicksham, O. Neill, B. R. Jones, and E. Smith. 2000. Global scale patterns of forest fragmentation. Conservation Ecology 4: 3. Robertson, B. A., and R. L. Hutto. 2006. A framework for understanding ecological traps and an evaluation of existing evidence. Ecology 87: 1075–1085. Rooney, N., K. S. McCann, and J. C. Moore. 2008. A landscape theory for food web architecture. Ecology Letters 11: 867–881. Rooney, T. P., and D. M. Waller. 2003. Direct and indirect effects of white-tailed deer in forest ecosystems. Forest Ecology and Management 181: 165–176. Saavedra, S., F. Reed-Tsochas, and B. Uzzi. 2009. A simple model of bipartite cooperation for ecological and organizational networks. Nature 457: 463–466. Sabatino, M., N. Maceira, and M. A. Aizen. 2010. Direct effects of habitat area on interaction diversity in pollination webs. Ecological Applications 20: 1491–1497. Saunders, D. A. 1998. The importance of networks for communication and application of ecological research. Surrey, Beatty & Sons, Chipping Norton, North South Wales, Australia. Saunders, D. A., R. J. Hobbs, and C. R. Margules. 1991. Biological consequences of ecosystem fragmentation: A review. Conservation Biology 5: 18–32. Saura, S. 2008. Evaluating forest landscape connectivity through Conefor Sensinode 2.2. In R. Lafortezza, J. Chen, G. Sanesi, and T. R. Crow [eds.], Patterns and processes in forest landscapes: Multiple use and sustainable management, 403–422. Springer Verlag, New York, New York, USA. Saura, S., and J. Torne. 2009. Conefor Sensinode 2.2: A software package for quantifying the importance of habitat patches for landscape connectivity. Environmental Modelling & Software 24: 135–139. Schick, R. S., and S. T. Lindley. 2007. Directed connectivity among fish populations in a riverine network. Journal of Applied Ecology 44: 1116–1126. Schlaepfer, M. A., M. C. Runge, and P. W. Sherman. 2002. Ecological and evolutionary traps. Trends in Ecology & Evolution 17: 474–480. Schoener, T. W. 1989. Food webs from the small to the large: The Robert H. MacArthur Award Lecture. Ecology 70: 1559–1589. Shah, V. B., and B. H. McRae. 2008. Circuitscape: A tool for landscape ecology. In G. Varoquaux, T. Vaught, and J. Millman [eds.], Proceedings of the 7th Python in Science Conference, 62–66, Pasadena, California, USA, 2008. SciPy2008, website http:http://conference.scipy.org/proceedings/SciPy2008/. Slobodkin, L. B. 2001. The good, the bad and the reified. Evolutionary Ecology Research 3: 1–13. 516 American Journal of Botany Solé, R. V., F. Bartumeus, and J. G. P. Gamarra. 2005. Gap percolation in rainforests. Oikos 110: 177–185. Solé, R. V., and J. M. Montoya. 2006. Ecological network meltdown from habitat loss and fragmentation. In J. A. Dunne and M. Pascual [eds.], Ecological networks, 305–323. Oxford University Press, Oxford, UK. Soulé, M. E., J. A. Estes, J. Berger, and C. M. Del Rio. 2003. Ecological effectiveness: Conservation goals for interactive species. Conservation Biology 17: 1238–1250. Soulé, M. E., J. A. Estes, B. Miller, and D. L. Honnold. 2005. Strongly interacting species: Conservation policy, management, and ethics. BioScience 55: 168–176. Spencer, M., and P. H. Warren. 1996. The effects of habitat size and productivity on food web structure in small aquatic microcosms. Oikos 75: 419–430. Srinivasan, U. T., J. A. Dunne, J. Harte, and N. D. Martinez. 2007. Response of compex food webs to realistic extinction sequences. Ecology 88: 671–682. Staddon, P., Z. Lindo, P. Crittenden, F. Gilbert, and A. Gonzalez. 2010. Connectivity, non-random extinction, and ecosystem function in experimental metacommunities. Ecology Letters 13: 543–552. Staniczenko, P. P. A., O. T. Lewis, N. S. Jones, and F. Reed-Tsochas. 2010. Structural dynamics and robustness of food webs. Ecology Letters 13: 891–899. Steffan-Dewenter, I. 2003. Importance of habitat area and landscape context for species richness of bees and wasps in fragmented orchard meadows. Conservation Biology 17: 1036–1044. Steffan-Dewenter, I., A.-M. Klein, V. Gaebelle, T. Alfert, and T. Tschamtke. 2006. Bee diversity and plant–pollinator interactions in fragmented landscapes. In N. M. Waser and J. Ollerton [eds.], Plant– pollinator interactions: From specialization to generalization, 387–407. University of Chicago Press, Chicago, Illinois, USA. Steffan-Dewenter, I., U. Munzenberg, C. Burger, C. Thies, and T. Tscharntke. 2002. Scale-dependent effects of landscape context on three pollinator guilds. Ecology 83: 1421–1432. Stouffer, D. B., and J. Bascompte. 2010. Understanding food-web persistence from local to global scales. Ecology Letters 13: 154–161. Sugihara, G., and H. Ye. 2009. Cooperative network dynamics. Nature 458: 979–980. Swift, T. L., and S. J. Hannon. 2010. Critical thresholds associated with habitat loss: A review of the concepts, evidence, and applications. Biological Reviews of the Cambridge Philosophical Society 85: 35–53. Terborgh, J., L. Lopez, P. V. Nuñez, M. Rao, G. Shahabuddin, G. Orihuela, M. Riveros, et al. 2001. Ecological meltdown in predator-free forest fragments. Science 294: 1923–1926. Tewksbury, J. J., D. J. Levey, N. M. Haddad, S. Sargent, J. L. Orrock, A. Weldon, B. J. Danielson, et al. 2002. Corridors affect plants, animals, and their interactions in fragmented landscapes. Proceedings of the National Academy of Sciences, USA 99: 12923–12926. Theobald, D. M. 2006. Exploring the functional connectivity of landscapes using landscape networks. In K. R. Crooks and M. Sanjayan [eds.], Connectivity conservation, 416–443. Cambridge University Press, Cambridge, UK. Tilman, D., R. M. May, C. L. Lehman, and M. A. Nowak. 1994. Habitat destruction and the extinction debt. Nature 371: 65–66. Trani, M., and R. Giles. 1999. An analysis of deforestation: Metrics used to describe pattern change. Forest Ecology and Management 114: 459–470. Treml, E. A., P. N. Halpin, D. L. Urban, and L. F. Pratson. 2008. Modeling population connectivity by ocean currents, a graph-theoretic approach for marine conservation. Landscape Ecology 23: 19–36. Tylianakis, J. M. 2009. Warming up food webs. Science 323: 1300–1301. Tylianakis, J. M., R. K. Didham, J. Bascompte, and D. A. Wardle. 2008. Global change and species interactions in terrestrial ecosystems. Ecology Letters 11: 1351–1363. Tylianakis, J. M., E. Laliberte, A. Nielsen, and J. Bascompte. 2010. Conservation of species interaction networks. Biological Conservation 143: 2270–2279. Tylianakis, J. M., T. Tscharntke, and O. T. Lewis. 2007. Habitat modification alters the structure of tropical host–parasitoid food webs. Nature 445: 202–205. Urban, D., and T. Keitt. 2001. Landscape connectivity: A graph-theoretic perspective. Ecology 82: 1205–1218. Urban, D. L., E. S. Minor, E. A. Treml, and R. S. Schick. 2009. Graph models of habitat mosaics. Ecology Letters 12: 260–273. Vázquez, D. P., and M. A. Aizen. 2003. Null model analyses of specialization in plant–pollinator interactions. Ecology 84: 2493–2501. Vázquez, D. P., and D. Simberloff. 2003. Changes in interaction biodiversity induced by an introduced ungulate. Ecology Letters 6: 1077–1083. Vellend, M., K. Verheyen, H. Jacquemyn, A. Kolb, H. Van Claster, G. Peterken, and M. Hermy. 2006. Extinction debt of forest plants persists for more than a century following habitat fragmentation. Ecology 87: 542–548. Vos, C., P. Berry, P. Opdam, H. Baveco, B. Nijhof, J. O’Hanley, C. Bell, et al. 2008. Adapting landscapes to climate change: Examples of climate-proof ecosystem networks and priority adaptation zones. Journal of Applied Ecology 45: 1722–1731. Wennergren, U., M. Ruckelshaus, and P. Kareiva. 1995. The promise and limitations of spatial models in conservation biology. Oikos 74: 349–356. Williams, S. E., L. P. Shoo, J. L. Isaac, A. A. Hoffmann, and G. Langham. 2008. Towards an integrated framework for assessing the vulnerability of species to climate change. PLoS Biology 6: e325. Wilson, K. A., E. Meijaard, S. Drummond, H. S. Grantham, L. Boitani, G. Catullo, L. Christie, et al. 2010. Conserving biodiversity in production landscapes. Ecological Applications 20: 1721–1732. Wright, D. H., A. Gonzalez, and D. C. Coleman. 2007. Changes in nestedness in experimental communities of soil fauna undergoing extinction. Pedobiologia 50: 497–503. Yaacobi, G., Y. Ziv, and M. L. Rosenzweig. 2007. Habitat fragmentation may not matter to species diversity. Proceedings of the Royal Society of London, B, Biological Sciences 274: 2409–2412. Zeigler, B. P. 1977. Persistence and patchiness of predator–prey systems induced by discrete event population exchange mechanisms. Journal of Theoretical Biology 67: 687–713.