* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Levin, S.A. (editor). Encyclopedia
Biogeography wikipedia , lookup
Conservation biology wikipedia , lookup
Storage effect wikipedia , lookup
Maximum sustainable yield wikipedia , lookup
Soundscape ecology wikipedia , lookup
Conservation movement wikipedia , lookup
Extinction debt wikipedia , lookup
Wildlife corridor wikipedia , lookup
Restoration ecology wikipedia , lookup
Biodiversity action plan wikipedia , lookup
Decline in amphibian populations wikipedia , lookup
Mission blue butterfly habitat conservation wikipedia , lookup
Occupancy–abundance relationship wikipedia , lookup
Habitat destruction wikipedia , lookup
Landscape ecology wikipedia , lookup
Reconciliation ecology wikipedia , lookup
Biological Dynamics of Forest Fragments Project wikipedia , lookup
Theoretical ecology wikipedia , lookup
Habitat conservation wikipedia , lookup
Provided for non-commercial research and educational use only. Not for reproduction, distribution or commercial use. This article was originally published in the Encyclopedia of Biodiversity, second edition, the copy attached is provided by Elsevier for the author’s benefit and for the benefit of the author’s institution, for non-commercial research and educational use. This includes without limitation use in instruction at your institution, distribution to specific colleagues, and providing a copy to your institution’s administrator. All other uses, reproduction and distribution, including without limitation commercial reprints, selling or licensing copies or access, or posting on open internet sites, your personal or institution’s website or repository, are prohibited. For exceptions, permission may be sought for such use through Elsevier’s permissions site at: http://www.elsevier.com/locate/permissionusematerial Pearson Scott M. (2013) Landscape Ecology and Population Dynamics. In: Levin S.A. (ed.) Encyclopedia of Biodiversity, second edition, Volume 4, pp. 488-502. Waltham, MA: Academic Press. © 2013 Elsevier Inc. All rights reserved. Author's personal copy Landscape Ecology and Population Dynamics Scott M Pearson, Mars Hill College, Mars Hill, NC, USA Published by Elsevier Inc. Glossary Core and satellite model Network of habitat patches in which centrally located core patches harbor stable populations that provide emigrants (or colonists) to ephemeral populations in peripheral satellite patches. Deme Localized breeding group or subpopulation, often confined to a single habitat patch. Effective population size (Ne) Number of individuals in population that participate in breeding. Extinction vortex Complex of interacting mechanisms that lead to the extinction of small populations. These mechanisms include loss of genetic diversity as well as environmental and demographic stochasticity. Habitat fragmentation Process in which habitat loss converts a landscape dominated by large connected patches Introduction Basic Concepts and Terminology about Population Dynamics The discipline of landscape ecology emphasizes the causes and consequences of spatial pattern on the functioning of populations, communities, and ecosystems. Nature is generally not homogenous with respect to ecological properties. Those properties include gradients in abiotic factors, such as the temperature, moisture, and the abundance of required resources; and biotic factors, such as the presence of predators, pathogens, and competing species. This spatial heterogeneity exists at multiple scales of space and time. The term ‘‘landscape’’ evokes patterns over a broad scale, such as the view from an airplane or satellite extending over several square kilometers and including the mosaic of forests, grasslands, wetlands, agricultural fields, and urban areas seen below (Figure 1). This mosaic and the diversity of habitats therein is the setting where species make their living. The spatial pattern of habitats determines how species and their populations are distributed across this view from above. The influence of spatial pattern is also important at finer scales. Studies in landscape ecology have illustrated how the spatial pattern of resources at the scale of 1 m2 altered the habitat use by beetles and ants (Wiens and Milne, 1989; Wiens et al., 1993). ‘‘Scale’’ is an important concept in landscape ecology because the influence of spatial patterns often depends on the scale in which they are observed and measured. It is common to measure spatial heterogeneity at multiple scales in order to best discern which scales are important for a given ecological phenomenon (e.g., Price et al., 2005). No matter what scale is studied, the spatial pattern of resources alternatively provides opportunities for population spread and increase, limits and constrains population growth, or threatens the persistence of some species. 488 of habitat to one in which habitat exists in small, disconnected patches or a condition in which habitat is spatially distributed as small, disconnected patches. Landscape matrix The complex of habitats occupying the space between habitat patches. Metapopulation Network of individual demes connected by the processes of immigration and emigration. Population dynamics Change in population size through time and the processes that bring about these changes. Sink population Deme in which death rates exceed birthrates. Sink populations will go extinct without an input of immigrants. Source population Deme in which birthrates exceed death rates. Sources produce an excess of individuals that emigrate from that deme. A population is a group of interbreeding organisms that produce viable, fertile offspring; under the biological species concept, members of a population belong to a single species. Localized breeding groups are also known as ‘‘demes.’’ In a typical landscape, the suitable habitat for most species exists in a patchy pattern (Figure 1), so patches of habitat usually define the location and boundaries of demes. Ecologists are interested in the number of individuals in a population (that is, population size classically denoted by the letter N). The change in N through time (dN/dt) is population dynamics. Population size changes by the demographic processes of birth, immigration, death, and emigration (BIDE). This BIDE model of population growth is dN=dt ¼ Births þ Immigrants2Deaths2Emigrants: Populations in which immigration and/or emigration occurs are considered to be open populations and are connected to other demes by the movement of individuals. The distances between demes influence the frequency of these deme-todeme movements. Movements may happen regularly for the population but occur infrequently during the life cycle of an individual. Movements often take the form of dispersal from the natal site to a different habitat patch. Dispersal lessens the probability of competition and interbreeding with parents and siblings, especially when the population in the natal patch is small. Analyses of population viability and genetic diversity emphasize effective population size (Ne), which is the number of individuals actually participating in breeding. In most populations, NeoN because some individuals are unsuccessful in attracting a mate, reproductively immature, or of postreproductive age. Encyclopedia of Biodiversity, Volume 4 http://dx.doi.org/10.1016/B978-0-12-384719-5.00417-2 Author's personal copy Landscape Ecology and Population Dynamics Aerial photograph 0 0.25 489 Classified land cover 0.5 1 Km Streams Roads Agriculture Wetland Mixed forest Hardwood Miscellaneous nonforest Conifer Developed Figure 1 Landscapes are a mosaic of habitat types. Color infrared aerial photograph from 1998 (left) and classification of land cover in 2001 (right) for a region of southern Appalachian Mountains, Madison County, NC, USA. Population Dynamics in Landscapes The number of individuals in a population, N, is limited by available resources. The limiting resource(s) may include food or nest sites for animals, or moisture, sunlight, and/or soil nutrients for plants. Suitable habitats are the locations and conditions that provide these resources. Given that resources are finite, N has a maximum value that is sustainable. Ecologists use the term carrying capacity (K) to refer to the maximum N that a given environment can support over the long term. The logistic growth equation is a simple model of population dynamics learned by most students of ecology: dN KN ¼ rN dt K The value dN/dt represents change in N over a unit of time (t), and r is the intrinsic rate of increase, which is a measure of per capita population growth: r¼ births deaths N Assumptions of this model include a closed population (i.e., there is no immigration or emigration) and that r and K are constants. Other ecological forces that could influence population growth, such as predation or competition, are not included. The logistic growth model implies that a population attains equilibrium with resources in its environment. The continuous form of this model produces graphs of N against time as an S-shaped curve in which the population grows until N ¼ K and then N remains constant at that value. The assumptions of the logistic growth model are rather restrictive, and very few if any populations in nature would fulfill all of them. Nevertheless, it expresses a central premise that populations are limited by availability of resources and the amount of habitat in the landscapes where they live. Furthermore, landscape ecologists study how the spatial arrangement of habitat patches can alter population dynamics. Landscape-level changes in habitat abundance and spatial pattern affect population dynamics. Data on populations of grassland birds in North America provide example of how changes in habitat availability alter population size. At the continental scale, the abundance of bird species that require grassland habitats has been declining since the mid-1900s in Author's personal copy 490 Landscape Ecology and Population Dynamics Regional response to habitat protection Population trends for two grassland birds (a) Meadowlark Field sparrow 30 25 20 15 10 5 Henslow’s sparrow Regional abundance (Birds/5000 hours) 50 45 40 35 30 25 20 15 10 5 0 1965 FISP abundance index EAME abundance index 250 1985 1995 2005 Year 150 100 50 0 1975 0 1975 200 (b) 1980 1985 1990 1995 2000 2005 Year Figure 2 Change in abundance for grassland birds at (a) continental scale for eastern North America and (b) regional scale for Illinois. (a) Data for Eastern meadowlark (Sturnella magna) and field sparrow (Spizella pusilla) provided by United States Geological Survey (USGS) North American Breeding Bird Survey, http://www.mbr-pwrc.usgs.gov/bbs/bbs.html. Solid line is mean estimate, with dashed lines showing 95% credible interval. (a) and (b) Reprinted from Herkert JR (2007) Evidence for a recent Henslow’s Sparrow population increase in Illinois. Journal of Wildlife Management 71: 1229–1233, with permission from Wiley. North America (Peterjohn and Sauer, 1999; Figure 2) and parts of Europe (Chamberlain and Fuller, 2001; Figure 2). These declines can be explained by the loss of grassland habitats because of changing agricultural practices, expansion of suburban development, conversion of grassland habitats to forests, and long-term disruption of ecological processes (e.g., fire and grazing) that create and maintain grassland habitats (Askins et al., 2007). Landscape-level analyses have revealed the specific habitat features required for foraging and nesting by these birds (Ribic and Sample, 2001; Veech, 2006). These species are sensitive to the fragmentation of their habitats. In response to these population declines, management initiatives at the regional scale have focused on habitats protection and restoration. One such initiative is the US Department of Agriculture’s Conservation Reserve Program (CRP). The CRP encourages the protection of large patches of grassland habitat and has the potential to increase habitat availability for these birds. Herkert (2007) documented an increase in Henslow’s sparrow (Ammodramus henslowii), an obligate grassland species, in response to the establishment of 400,000 ha of CRP lands in Illinois (Figure 2) but cautioned that local management practices (such as farmers’ mowing schedules) can affect habitat quality. The positive effects of similar habitat restoration efforts have been documented in other regions (e.g., Benson et al., 2006). Monitoring population dynamics at the landscape scale is challenging. Various survey techniques are available for estimating population size, but measurement of demographic rates is more difficult. It is generally necessary to estimate demographic rates (i.e., births, deaths, and immigration) by sampling a limited number of sites or habitat patches in the landscape because of the large effort required to produce these estimates. Often it is the variation in demographic rates among habitat patches that is of greatest interest. For example, Herkert et al. (2003) sampled 39 sites to estimate the effects of patch size on nest survival and parasitism in three grassland bird species. In a study of song sparrow (Melospiza melodia) populations, Jewell and Arcese (2008) linked the population growth rates of the sparrows indirectly to landscape-level patterns of land use (Figure 3). Rates of brood parasitism by cowbirds (Molothrus ater) reduced sparrow reproduction, and cowbird abundance was strongly correlated with land uses such as urban and agricultural lands with livestock grazing. These land uses created areas where parasitism rates were so high that local reproduction in sparrow populations could not replace individuals lost to mortality (i.e., ro0). Their results suggest that these populations were maintained by immigration from areas with lower parasitism rates rather than by local reproduction. Simulation models are often employed to explore the effects of landscape patterns on population dynamics. Spatially explicit models can simulate the occupancy of specific locations as well as the demographic rates characteristic of populations at those locations. Such models allow the researcher to test hypotheses about population dynamics in the complex settings of real and theoretical landscapes. These models are especially useful for studying the impacts of various scenarios of landscape change on populations. Pearson et al. (1999) examined how management options related to forest clearing would affect habitats, and therefore the abundances, of a suite of native species. Complex population processes such as survival, reproduction, and movement of individuals can be represented in spatially explicit models. For example, Holdo et al. (2011) used a model to study the impact of road construction that would interfere with migratory movements of the Serengeti wildebeest (Connochaetes spp.; Figure 4). Their simulations suggest that this barrier to movement could reduce wildebeest population by one-third because these animals would be less able to migrate to the most productive foraging sites. Locations of the most productive sites shift in location seasonally, and the road interfered with the animals’ ability to track and migrate to those sites. Moreover, the road was predicted to divide a single wildebeest population into two separate subpopulations. Author's personal copy Landscape Ecology and Population Dynamics Brood parasitism rates 491 Population growth Expected population growth rate Parasitism rate 0⋅0−0.2 Low 0⋅2−0.4 Moderate 0⋅4−0.8 High < 0⋅90 Strong sink 0⋅90−1⋅00 Weak sink 1⋅00−1⋅10 Weak source No song sparrows N 012 4 6 8 (a) N 012 4 6 8 (km) (b) (km) Figure 3 Maps of brood parasitism and population growth of song sparrows (Melospiza melodia) in the Southern Gulf Islands of British Columbia, Canada. Rates of parasitism by brown-headed cowbirds (Molothrus ater) (a) are affected by land use and results in variation in population growth rates of their sparrow hosts (b). Growth rates are expressed in terms of finite rates of increase. Rates Z1 indicate stable or growing populations (sources), whereas rates o1 represent declining or sink populations. Reprinted from Jewell KJ and Arcese P (2008) Consequences of parasite invasion and land use on the spatial dynamics of host populations. Journal of Applied Ecology 45: 1180–1188, with permission from Wiley. 1600 Wildebeest (1000s) 1° S Lake Victoria 1200 Wildebeest dry season Nairobi Musoma 800 Kenya 400 Tanzania 0 0 20 40 60 Time (yr) Total population (no road) Total population (road) 80 Wildebeest wet season 3° S 100 Northern subpopulation (road) Southern subpopulation (road) Arusha 33° E Migration route Proposed road Figure 4 Potential impact of a proposed road on Serengeti wildebeest populations. (a) Results of spatially explicit simulation model on abundance of wildebeest (Connochaetes spp.) comparing the landscapes with (red line) and without (black line) proposed road. The road would separate wildebeest into northern and southern populations shown by blue and green lines. Graph reprinted from Holdo RM, Fryxell JM, Sinclair ARE, Dobson A, and Holt RD (2011) Predicted impact of barriers to migration on the Serengeti wildebeest population. Public Library of Science One 6: 1–7. Map provided by Stuart Pimm, with permission from PLOS. Although the overall abundance of required resources is of great importance for populations, the patchiness and spatial arrangement of habitat types have the potential to complicate population dynamics in landscapes. Demographic rates, such as survival and reproduction, may be affected by patch size and shape. Movements between patches can be influenced by the distance between patches and other spatial features. Therefore, an understanding of how population dynamics play out in space is important for research and management. Author's personal copy 492 Landscape Ecology and Population Dynamics Landscape Structure and Population Dynamics Patches, Matrix, and Corridors Landscape ecologists are interested in spatial patterns. For populations, the overall abundance of habitat and its spatial distribution have a strong influence on the genetic structure and population dynamics of species. The concepts of patch, matrix, and corridor are useful for understanding how populations are influenced by the spatial structure of landscapes (Figure 5). For the time being, assume that patches are recognizable as discrete areas of similar habitat and that all habitat patches contain suitable habitat with enough resources to sustain a population. (This assumption of discrete, homogeneous patches will be relaxed later.) The unsuitable habitats found between habitat patches comprise the landscape matrix. Corridors are linear strips of habitat that connect patches and facilitate the movements among patches (Figure 5). The number, sizes, shapes, and spatial arrangement of habitat patches, as well as the presence of corridors, impose a spatial structure on the species that occupies them. Several terms contain the word ‘‘structure.’’ The spatial distribution of habitat may be referred to as ‘‘patch structure’’ when applied to a specific habitat type or as ‘‘landscape structure’’ when applied to the spatial pattern of all habitat types in the landscape. The term ‘‘population structure’’ refers to the spatial distribution of breeding groups (i.e., demes) and the rates of exchange of individuals, via emigration and immigration, among these groups. Much of the problem of analyzing population dynamics at the landscape level relates to understanding the influence of landscape (or patch) structure on population structure. The patch structure of a landscape influences the dynamics of populations living therein because patch structure affects the spatial distribution of individuals and their movements (Saunders et al., 1991). Patch size is one of most important qualities of landscape structure because it is directly related to local population size. Larger patches have a greater abundance of resources and will support a greater N over the long term (i.e., the patch’s carrying capacity), whereas smaller patches support fewer individuals. Immigration and emigration are affected by the distance between patches and the intervening landscape matrix. When patches are far from one another, the chance of successful movement between them is reduced. Dispersing animals can suffer mortality from hazards of predation, starvation, or accident while crossing the unsuitable matrix. The dispersal mechanisms of plant seeds may fail to deliver them to distant patches, and dispersing animals may not find a suitable patch. Corridors connect patches and facilitate interpatch movements. Similarly, small habitat patches (i.e., habitat fragments) scattered throughout the matrix can act as ‘‘stepping stones’’ to promote movement between large patches. Interpatch distances between these fragments are small, and the habitat fragments provide temporary refuge for the dispersing animal before crossing the next gap. Likewise, these fragments may be too small to support a plant population over the long term, but temporary populations can produce seeds capable of colonizing nearby patches. The effectiveness of corridors and stepping stones for movement can be altered by characteristics of the landscape matrix (Baum et al., 2004; Vergara, 2011). Satellite patches Corridors Core patch Satellite patches Stepping stones Habitat patches Landscape matrix Figure 5 Schematic diagram of patch–matrix–corridor model. In this example, a centrally located core patch is surrounded by smaller satellite patches. Individuals from the population of the core patch disperse to the satellite patches, which support smaller populations that frequently go extinct. Linear corridors or closely spaced habitat fragments (stepping stones) facilitate movement among patches. Author's personal copy Landscape Ecology and Population Dynamics Characteristics of the landscape matrix may facilitate or hinder movement. Habitats in the matrix may be harsh or benign relative to the resource needs of individual species. The matrix may contain barriers to movement. The barriers include natural features such as rivers or mountains and human constructed features such as roadways, agricultural fields, and urban areas. The relative ease or difficulty of movement through the landscape matrix is called its permeability. The spaces between habitat patches, filled by the matrix, are called gaps. A species’ ability to move across this space is called its gap-crossing ability. Gap-crossing ability and matrix permeability for a given species in a particular landscape depend on the life-history characteristics, habitat needs, and vagility of the species as well as the matrix itself (Chardon et al., 2003; Baguette and Van Dyck, 2007). For example, a bird can readily fly across 100-m wide plowed agricultural field, but that same field is a barrier to a salamander, which risks predation or death by desiccation during an attempt to cross it. Landscape Context The suitability and uses of particular habitat patches may be affected by landscape context. The matrix and landscape-level abundance of similar habitats alter the abundance of a species at the patch scale (Radford and Bennett, 2007). When a given habitat type is abundant in the broader landscape, then it is more likely to see occupancy or visits to small patches (Pearson, 1993; Lewis et al., 2011). This effect has been documented in mobile animals, such as mammals, flying insects, and birds, which readily move between patches. Landscape context affects plant population dynamics by altering pollination services and rates of seed predation (Steffan-Dewenter et al., 2001). The abundance and diversity of habitats in the broader matrix affect the probability that pollinators or seed predators will visit a given patch. Some animal species depend on a variety of habitat types during their lives and the ability to move among them. These species may be especially sensitive to landscape context. For example, some amphibians require wetlands for breeding but then disperse to upland habitats after the breeding season (Pillsbury and Miller, 2008). Such species require different habitat types in different seasons or different parts of their life cycles. Therefore, a mix of patch types that provide supplementary or complementary resources (sensu Dunning et al., 1992) must be considered when assessing habitat quality and quantity at patch and landscape scales. By altering rates of immigration and emigration, landscape context can affect the genetic diversity and divergence among populations (Blevins et al., 2011). Quantifying Landscape Structure and Connectivity Spatial metrics provide a means of quantifying the spatial arrangement of habitat. The most straightforward metrics focus on the statistical distribution of patch sizes. First, the landscape is classified into suitable versus unsuitable habitat (i.e., a binary landscape), and the areas of suitable patches are tallied. Then the ‘‘number of patches,’’ ‘‘mean patch size,’’ and ‘‘size of the largest patch’’ are useful for comparing landscapes with similar amounts of habitat but having a different spatial arrangement of habitat. Other metrics can measure more complex aspects of spatial pattern such as the shape of patches. These metrics incorporate a ratio of perimeter-to-area for 493 patches. Compact shapes, such as circular or square patches, have a lower perimeter-to-area ratio, whereas elongated patches or those with complex shapes have a greater perimeter-toarea ratio. The metric of ‘‘contagion’’ (e.g., Riitters et al., 1996) measures the intermixing of habitat types, which is correlated with number of patches as well as patch size and shape. There are many spatial metrics available, and many are correlated with each other (Hargis et al., 1998). No single metric will measure all of the important aspects of spatial pattern. The best choices will be those metrics that measure the qualities of landscape structure predicted to influence the population processes under study. One of the most important qualities is habitat connectivity, which measures the ease of moving between habitat patches. Landscape ecologists have developed a variety of metrics for measuring the connectivity of landscapes (Kindlmann and Burel, 2008). These metrics incorporate the interpatch distances, area of habitat at various distances from the focal habitat, and the presence of stepping stones and corridors (Leidner and Haddad, 2011). A proximity index rates the relative amount and distance to suitable habitat for single patches (Gustafson and Parker, 1994). Deciding which among the available metrics is the best choice for any given species remains challenging. Furthermore, connectivity has been treated as both independent and dependent variables in landscape ecology research (Goodwin, 2003). When considered an independent variable, connectivity was considered a structural aspect of the landscape under study. These studies often compared the performance of populations living in landscapes having different patch structures. When treated as a dependent variable, connectivity was a function of various species responses to pattern in the same landscape. A highly mobile species, such as a bird or large mammal, can easily move between patches, but a sedentary organism or nonflying invertebrate will find the same landscape disconnected (or fragmented) if it cannot readily move among patches. Thus, sensitivity to habitat loss and fragmentation depends on species’ habitat needs and life-history traits (Driscoll and Weir, 2005; Sutton and Morgan, 2009). Landscape Structure, Population Viability, and Conservation Conservation biology naturally considers the effects of landscape structure on population structure and population viability. Therefore, there is a close and complementary connection between the disciplines of Conservation Biology and Landscape Ecology. The conservation of any species depends on the maintenance of viable populations, and the patch structure of landscapes affects population viability at both the patch and landscape level. Population Viability in Small Patches Population viability is strongly related to effective population size (Ne) and genetic diversity, two characteristics that are linked. Small populations suffer from demographic stochasticity and reduced genetic diversity. Populations in landscapes Author's personal copy 494 Landscape Ecology and Population Dynamics in which small isolated patches predominate show less genetic diversity overall than populations in unfragmented landscapes (e.g., butterfly (Collier et al., 2010) and toad (Allentoft et al., 2009)). Moreover, population sizes in small patches fluctuate to a greater degree because of demographic stochasticity, which is the fluctuation in Ne, birthrates, and death rates through time. They also fluctuate as a result of environmental stochasticity, which is variation in demographic rates from unpredictable events related to factors such as weather. Such fluctuations create a greater chance of extinction and expose the population to loss of diversity because of genetic drift and population bottlenecks. The viability of populations in small patches is reduced because of limited genetic diversity. Small populations contain less genetic diversity because there are fewer individuals. In addition, genetic drift and inbreeding further reduce genetic diversity over time. Genetic drift is the random change in allele frequencies because of sampling error in the process of mating. Over the long term, drift ultimately leads to the loss of alleles when their frequencies randomly reach zero. Levels of heterozygosity, a common way of measuring genetic diversity, are positively associated with population growth and persistence (Hanski and Saccheri, 2006). Individuals with higher levels of heterozygosity typically have higher survival and reproductive rates than individuals with less heterozygosity (Coltman et al., 1998). This phenomenon is called heterozygote advantage or heterosis. Heterozygosity declines as alleles are lost and the population becomes fixed for one allele. Similarly, inbreeding reduces heterozygosity among offspring when related individuals breed, and inbred populations have a greater chance of extinction (Saccheri et al., 1998). Inbred offspring have lower levels of heterozygosity and have a greater chance of being homozygous for deleterious recessive alleles. This reduced fitness is termed ‘‘inbreeding depression.’’ In small populations, the choice of mates is directly related to Ne, and matings among related individuals will inevitably occur. The primary conclusion is that small populations have lower genetic diversity and stochastic processes of genetic drift, inbreeding, and population bottlenecks will further reduce genetic diversity over time. The predominance of small patches in fragmented landscapes can affect demography and behaviors. Small populations may suffer reduced population growth rates because of social factors called Allee effects. Allee (1931) suggested that some species have a minimum viable population (MVP) size. Birth and death rates are density-dependent, and recruitment drops to unsustainable levels below a certain N. For example, fertilization rates in flowering plant may depend on visitation rates by pollinating insects, but a minimum density of individual plants is needed to attract the necessary numbers of pollinators (Hackney and McGraw, 2001). When Ne or densities are too low, recruitment rates decline because of low fertilization rates. Likewise, vulnerability to predators can increase in some animal groups. Flock size and herd size are correlated with the ability to detect predators and thereby reduce predation risk for the individual (Lima and Dill, 1990). Allee effects are apparent when survival and reproductive rates are positively correlated with population density and density is below the threshold for density-dependent population regulation (i.e., the carrying capacity). However, care should be taken not to confuse these effects with the negative effects of small population size such as inbreeding (Stephens et al., 1999). In fragmented landscapes, the small patches may support too few individuals needed to accrue the positive benefits of population size and/or density. For animals, patterns of behaviors such as movements can be altered by habitat fragmentation. For example, Mitrovich et al. (2009) found differences in the home range overlap and movements and crowding for the coachwhip (Masticophis flagellum), a large snake, in small patches relative to large patches. Movements of large highly mobile mammals may be affected. too. Seasonal movement behaviors of white-tail deer (Odocoileus virginianus) varied with the mean size and number of patches in a landscape in the midwestern USA (Grovenburg et al., 2011). Deer in landscapes with many large patches tended to be sedentary residents, whereas those in landscapes with small patches were more likely to be migratory. The genetic and demographic problems of small populations interact synergistically in a phenomenon referred to as an extinction vortex (Gilpin and Soule, 1986). When Ne falls, genetic diversity is lost because of the phenomena of bottlenecks, drift, and inbreeding. Reduced genetic diversity causes drops in population growth, reducing mean fitness because genetic diversity is correlated with the levels heterozygosity in individuals and the population as a whole. In small populations that have lost a significant amount of genetic diversity, low or negative growth rates further reduce N, which results in greater loss of genetic diversity. Moreover, reduced genetic diversity limits the ability of the population to adapt, via natural selection, to changes in its environment. Therefore, small populations are more sensitive to environmental change than large populations having more diversity. This combination of interacting processes eventually threatens the small population with extinction. What population size qualifies a small population? Although the most accurate answer will depend on the life history of specific species, the ‘‘50/500’’ rule of thumb from conservation biology (Franklin, 1980) provides a useful starting point when detailed life-history information is lacking. This rule states that isolated demes with an effective population size (Ne) greater than 500 will be relatively safe from the dangers of the loss of genetic diversity. These risks progressively increase as Ne declines from 500 toward 50. At Ne o 50, all combined forces of the extinction vortex act on the population, and extinction may be inevitable without management intervention. Thus, a Ne450 is needed for short-term population viability and Ne4500 needed for long-term viability. The ‘‘50/500’’ rule was based on goal of maintaining genetic diversity (Franklin, 1980). Shaffer (1981) outlined the concept known as the minimum viable population. Estimates of a MVP include three parameters: a Ne, a probability of extinction (or persistence), and a specific time period. For example, ‘‘An MVP for any given species in any given habitat is the smallest isolated population having a 99% chance of remaining extant for 1000 years despite the foreseeable effects of demographic, environmental, and genetic stochasticity, and natural catastrophes’’ (Shaffer 1981, p. 132). When incorporating the dangers of demographic and environmental stochasticity, the minimum Ne should be in the range of thousands rather than hundreds (Traill et al., 2010). Author's personal copy Landscape Ecology and Population Dynamics Some scientists criticize the notion of general guidelines for MVPs, arguing that the minimum Ne for long-term persistence must be determined on a species-by-species basis (Flather et al., 2011). Population viability and MVP will be affected by the unique combination of species’ traits, including its phylogeny, life history, resource needs, habitat availability, population structure and demography, and environmental threats (both natural and human caused) within the context of the landscape where it lives. For landscape-level conservation efforts, it should be more profitable to identify and address the specific causes of population declines (Caughley, 1994) rather than focus on maintaining a minimum number of individuals. Nevertheless, an estimate of MVP can be useful for labeling species as being at risk of extinction, motivating policy makers and managers to take appropriate actions, and setting a target population size for restoration efforts. Immigration and Rescue Effects Despite the problems described in the previous section, small populations are frequently observed in nature and seem to persist over the long term. Their persistence can be explained by the fact that these populations are open and connected to other demes in the landscape via the movement of individuals. Population biologists commonly refer to these movements as ‘‘immigration’’ (into) and ‘‘emigration’’ (from) when in reference to a specific deme. Immigration has several benefits to a small population and some risks. Immigrant individuals can increase genetic diversity if they bring alleles that are absent or rare in the deme. Even infrequent or low levels of immigration are sufficient to restore genetic diversity lost to genetic drift (Slatkin, 1985). Immigrants provide a form of gene flow. The interpatch transport of gametes, such as pollen for flowering plants, is another example of gene flow. Moreover, immigrants provide a form of population growth when local recruitment rates are low (see BIDE model in the section Basic Concepts and Terminology about Population Dynamics). Thus, immigration can periodically rescue a population from the negative effects of demographic stochasticity and loss of genetic diversity, but there are significant risks associated with the movement of individuals. ‘‘Outbreeding depression’’ occurs when immigrants introduce genetic traits that are poorly suited for the local environmental conditions. Offspring expressing these traits will have reduced fitness, and the mean fitness of the population will be lower until natural selection eliminates these traits. If immigration of these traits persists, then the mean fitness and population growth rate will be determined by a balance between immigration and selection. A second risk is the introduction of pathogens and parasites carried by immigrants. Preserving Habitat at Landscape Level: The SLOSS Debate Conservation biologists have debated the best strategy for preserving habitat for threatened and endangered species. If the total area of habitat that can be protected is limited, then what is the best scheme for distributing that habitat in space? The essence of this question is whether to protect the same 495 acreage of habitat in ‘‘single large or several small’’ (i.e., the acronym SLOSS) patches. There are advantages and disadvantages for both choices (Diamond et al., 1976). A single large preserve provides habitat to support the largest possible population. If Ne is sufficiently large, then this choice will avoid the problems of small populations discussed previously. However, the danger of a single habitat patch is that some environmental catastrophe such as a hurricane or fire or an introduced virulent pathogen could extirpate the entire species. If the species is spread across several smaller patches sufficiently separated in space, then the chance of a catastrophic event affecting all of them is less than if they are confined to a single patch. For the ‘‘several small’’ strategy to be viable, the demes of respective patches must be large enough to avoid the extinction vortex, or the populations must be open to movement between patches in order to accrue the benefits of immigration. The movement of immigrants may be natural or facilitated by human management in the form of translocations of individuals. Land management around preserved patches may include the preservation of corridors to connect patches and facilitate movements among them. Corridors likewise have their relative advantages and disadvantages (Haddad et al., 2011), and their effectiveness depends on the natural history of the target species (Gilbert-Norton et al., 2010). Species that are dispersallimited or have difficulty crossing the landscape matrix receive the greatest benefits corridors. For animals, corridors of lowquality habitat can enhance movement among habitat patches even if populations cannot persist within the corridor itself (Haddad and Tewksbury, 2005). Landscape Ecology and Metapopulation Dynamics Most populations in nature have a patchy distribution because of patchiness of their habitat, and population structure is determined by the exchange of individuals (e.g., immigration and emigration) among patches. If the rates of exchange are high and frequent, such as by a highly mobile species, then there is simply one population occupying multiple patches. If exchanges among patches are very infrequent or nonexistent, then each patch has a distinct and separate population. An intermediate level of exchange would include regular movements among patches by dispersing individuals moving from their natal site to another patch. This amount of exchange characterizes metapopulations. The dynamics of open populations persisting in a patchy landscape often take the form of a metapopulation. A metapopulation is a network of individual demes connected by the processes of immigration and emigration. Metapopulation dynamics exist when species have demes in discrete habitat patches. Each deme has a finite probability of extinction, but extinction events are not synchronized in time across space. Empty patches are common, but none of them are too isolated to be recolonized from an extant deme. Thus, a metapopulation exhibits turnover in the occupancy of patches that is a function of local extinctions and recolonizations (Hanski, 1999). McCullough (1996) provides examples of metapopulation ecology applied to problems of wildlife conservation. Author's personal copy 496 Landscape Ecology and Population Dynamics Metapopulation theory has been an important tool for helping ecologists think about population dynamics in fragmented landscapes, but detecting metapopulations and parameterizing metapopulation models from field data is still challenging (Driscoll, 2007). The incidence function model of (Hanski et al., 1996b) has been one of the most useful because it requires only patch occupancy data. By monitoring how many and which patches are occupied over time, ecologists can relate the probability of occupancy to patch size, shape, and isolation. Alternative techniques include analyses of genetic structure to infer the relative isolation of habitat patches and the frequency of immigration among them (Wang et al., 2008). Sources and Sinks In the simplest case of population structure, the size of patches and the quality of habitat contained within them is similar, but this case is likely uncommon in nature. In a metapopulation spread over several habitat patches, variation in population growth rates can exist because of patch-specific differences in habitat quality, interspecific interactions, or other ecological factors. Pulliam (1988) described the situation in which some patches have intrinsic rates of increase (r ¼birthrate death rate) greater than 0 – that is, local birthrates exceed local death rates. These patches were termed ‘‘sources’’ because local population growth will produce an excess of individuals. The local population growth exceeds the patch’s carrying capacity, and these excess individuals must emigrate; if not, then feedbacks from local resource limitation will reduce growth rates. In the same metapopulation, other patches called sinks have ro0 because the local birthrate is less than the death rate. If isolated from other populations, extinction is inevitable for sink populations. However, sinks can persist in the context of a metapopulation because excess individuals from source patches can immigrate into sinks. These sinks receive a demographic subsidy and are able to persist because the sum of (births þ immigrants deaths)Z0. The knowledge of source and sink patches is critical for understanding how a particular metapopulation functions. This understanding is important for landscape-level conservation efforts in which choices are made about the protection (or allowing the destruction) of some fraction of the existing patches. If a critical number of the sources patches are lost, then the entire metapopulation will collapse (Hanski and Ovaskainen, 2002). Unfortunately, patch occupancy alone is not a reliable indicator of the source/sink status of individual patches. The hard work of estimating demographic rates must be undertaken to fully understand this aspect of population dynamics. Core and Satellite Model In conservation planning, the ‘‘core and satellite model’’ of habitat protection has been used at the landscape level (Shen et al., 2008; Snep and Ottburg, 2008). This model envisions critical habitat distributed in patches varying in area and quality. Core patches support the largest Ne and populations that persist through time. Core populations may act as demographic sources, and their greatest value to the broader metapopulation is to produce emigrants destined for satellite populations (Runge et al., 2006). Small patches located near the periphery of the core patches support small satellite populations that periodically go extinct. Satellite populations depend on immigrants from the core patches to avoid extinction or for recolonization after local extinction. Satellite populations may be demographic sinks or simply have a small Ne that is vulnerable to the problems of small populations. Landscape Change, Habitat Fragmentation, Edge Effects, and Extinction Debt Landscape changes, particularly those caused by human land uses, can alter the abundance and connectivity of suitable habitat. Habitat fragmentation results when habitat loss converts a landscape dominated by large connected patches of habitat to one in which habitat exists in small, disconnected fragments. Fragmentation of native habitats is happening on a global scale (Riiters et al., 2000). This phenomenon affects habitat quality as well as its quantity and spatial distribution. Consequentially, fragmentation affects the spatial structure of populations. The effects of fragmentation can be conceptually separated into the effects of (1) reduced patch size and (2) increased patch isolation. The processes of habitat loss and fragmentation often occur concurrently, but Fahrig (2003) argued that their effects need to be considered independently to best understand the consequences of landscape change on populations and communities. In fragmented landscapes, the effects of small patch size are confounded with edge effects. Small fragments have more edge, which can alter habitat quality. Small habitat patches have a greater edge-to-interior ratio because of simple geometry. As the area of a circle (p radius2) decreases, the perimeter of a circle (2 p radius) decreases more slowly. Circles and squares are simple, compact shapes in which length and width are approximately equal. Habitat fragments of elongated and complex shapes have a greater amount of perimeter compared to the area. The patch perimeter or ‘‘edge’’ is the boundary between the habitat of the patch and the landscape matrix, and the ecological conditions at the patch edge can be modified by the conditions in the matrix (Matlack, 1993). For example, consider a forest patch in a matrix of agricultural fields. The edges of forest are drier, warmer, windier, and receive more sunlight than locations deep in the interior of the forest patch. These altered conditions are termed ‘‘edge effects’’ and may extend some distance into the forest patch. Edge effects also include differences in the biotic interactions. For example, predation rates may be greater near edges. The ‘‘interior habitat’’ – sites within the patch but some minimum distance from the patch boundary – is considered free of edge effects. Some species are sensitive to the differences between edges and interiors. So-called ‘‘edge-sensitive’’ or ‘‘interiordependent’’ species avoid edges or experience lower survival and/or reproductive rates in edges. As fragment size decreases, the edge-to-interior ratio increases; edge-sensitive species have less and less interior habitat for use. Below a particular patch size, interior habitat is absent because edge effects penetrate to all regions of the small patch. Therefore, interior-dependent Author's personal copy Landscape Ecology and Population Dynamics species are absent from the smallest patches that are dominated by edge-tolerant species (Banks-Leite et al., 2010). Although the example of forest habitat was discussed here, edge effects also exist for grasslands and other nonforest habitats (Renfrew et al., 2005). The response of populations to the process of habitat loss and fragmentation may lag behind the changes in habitat quantity and quality. For example, some edge-sensitive species may persist in habitat fragments in spite of the loss of interior habitat. This lag occurs because individuals survive in small patches but long-term recruitment rates drop below mortality rates. Thus, small patches are occupied and some reproduction may be occurring, but not enough for long-term population persistence. In addition, the metapopulation dynamics may be disrupted by the overall loss of participating patches or the loss of a critical patch, and the metapopulation collapses (Hanski et al., 1996a). Eventually, these species will go extinct in this landscape, but extinction is delayed by the presence of long-lived individuals and some low levels of reproduction. This delay is termed ‘‘extinction debt’’ (Hanski and Ovaskainen, 2002; Honnay et al., 2005), and it depends on the life history of species affected by habitat fragmentation (Piqueray et al., 2011). Krauss et al. (2010) reviewed data for a variety of species living in fragmented grasslands of Europe. They found that long-lived perennial plant species were more likely to exhibit extinction debt than short-lived, mobile species such as butterflies. For species exhibiting extinction debt, their distribution on the landscape is better explained by past habitat patterns than by the distribution and quality of habitat in the contemporary landscape. Environmental Gradients and Habitat Heterogeneity at Multiple Scales The concept of source and sink populations (Pulliam, 1988) raised awareness of how variation in demographic performance among patches affects the occupancy of individual patches as well as the landscape-wide distribution of a population. The spatial arrangement of patches is important because sink patches may persist only if they are within the dispersal window of at least one demographic source (Foppen et al., 2000). In the original models of source and sink metapopulations, within-patch conditions were considered homogeneous, but this situation is seldom typical in real landscapes. Spatial heterogeneity within patches may affect demographic rates and population dynamics at fine scales. For large patches, spatial heterogeneity in habitat quality may create source–sink dynamics among sites within the patch. Environmental heterogeneity can create a mosaic of sites having higher and lower rates of survival, reproduction, and recruitment (Meekins and McCarthy, 2001). Within-patch heterogeneity can be especially high when patches are defined by a coarse-level criterion, such as land-cover type, and finescale heterogeneity is ignored. Fine-scale heterogeneity in temperature, moisture, soil nutrients, etc. may be correlated with occurrence of individuals (Palmer, 1990; Fraterrigo et al., 2006) and demographic rates (Meekins and McCarthy, 2001). If habitat quality is strongly correlated with fine-scale edaphic conditions (or some other measure of resource abundance for 497 such food availability), then these conditions may be used to estimate the demographic potential of a given site (Seydack et al., 2000; Ecke et al., 2002; Gonzalez-Megias et al., 2005; Flinn, 2007). Although not separated into discrete patches, source-sink dynamics can occur among the sites within patches. The demographic rates associated with an entire patch will be a function of the average habitat quality of its embedded sites. Landscape ecologists, with their appreciation for issues of scale, are well-equipped to study the relative effects of within-patch, among-patch, and landscape heterogeneity on population dynamics. Changing patterns of biodiversity in changing landscapes illustrate how spatial patterns at different scales interact with population processes. In portions of Europe and North America, forests are expanding following agricultural abandonment. New forest patches are developing from abandoned farmland via the process of secondary succession; this new habitat is also called postagricultural forest. The spatial arrangement of old forest patches is of great importance for native herbaceous plant species that depend on forest habitats. At the landscape scale, the abundance and proximity of ‘‘ancient forests’’ (not cleared for agriculture) is correlated with the patterns of plant diversity among recovering forest patches (Matlack, 1994a, b). The ancient forests preserved a pool of forest plant species during the time period of heavy agricultural use in the landscape (Pearson et al., 1998). During the period of recovery after the peak of agriculture, these species are able to disperse from the ancient forests to the new forest patches. Moreover, life-history traits influence the establishment and persistence of populations (Vellend, 2003; Flinn and Vellend, 2005) in particular patches. Verheyen et al. (2003) studied the distribution of forest herbaceous plant species in these postagricultural forests and their landscapes. They found that recovery after disturbance is a two-stage process in which the dispersal and colonizing ability of a species is initially most important in establishing populations in new forest patches. Next, after a species had colonized a new forest patch, habitat quality and niche specialization of the species influenced abundance and persistence within patches (also Pearson and Fraterrigo, 2011). The distribution of dispersal limited plant species was not correlated with habitat quality at the landscape scale because these species simply could not reach isolated forest patches. In other words, habitat fragmentation negated any effects of varying habitat quality for these postagricultural forests in the initial phase of recovery. Habitat quality became important only if and when a species reached a patch. This process takes many years to play out and will affect forest biodiversity for a century or more of recovery (Vellend et al., 2007). Continuous Habitat Variability in Landscapes The patch–matrix–corridor approach to characterizing spatial pattern is useful for landscapes in which patches exist as discrete readily identifiable units, and there are greater differences among patches than within them. However, in many landscapes, habitat variation occurs along environmental gradients that change relatively slowly across space (Figure 6). In such a landscape, drawing a definite boundary to denote habitat Author's personal copy 498 Landscape Ecology and Population Dynamics 0 Contours 125 250 500 m Drier Relative soil moisture Streams Wetter Figure 6 Map of predicted soil moisture as an example of continuous variation in habitat quality. Soil moisture is influenced by topography shown by 10-m contour lines as well as slope, aspect, and surrounding landforms. Drier sites are found on ridges and peaks, whereas wetter sites are located in ravines and valleys. This map was developed using a digital elevation model to calculate a topographic relative moisture index (Parker, 1982). (Reproduced with permission from Simon SA, Collins TK, Kauffman GL, McNab WH, and Ulrey CJ (2005) Ecological zones in the Southern Appalachians: First Approximation. Asheville, NC: US Department of Agriculture, Southern Research Station. Forest Service Research Paper SRS-41). Map shows 50-m grid cells. patches is nearly impossible because suitability varies in a complex and continuous manner across space. In mountainous landscapes, topography creates gradients in temperature, moisture, and light related to elevation, aspect, slope, and terrain shape (Figure 6). Instead of maps of discrete patches of suitable and unsuitable habitat, these landscapes are best represented by maps showing continuous variation in habitat quality (Thompson et al., 2006; Franklin, 2009). Quantifying the spatial pattern of habitats in these landscapes is challenging (Dorner et al., 2002). The statistical characteristics of this continuous variation can be quantified using a range of statistical techniques including blocking, spectral analysis, measures of spatial autocorrelation (Turner et al., 1991), or surface metrics (McGarigal et al., 2009). Patterns of habitat heterogeneity can change between fine and coarse scales, and these analyses can identify the particular spatial scale(s) in which spatial pattern and population processes are linked (Fortin and Dale, 2005). To be useful, the analysis should be approached from an organism-based perspective; that is, landscape structure can be understood by relating environmental gradients to the tolerances and resource needs of individual species (Cushman et al., 2010) and to population processes such as dispersal ability and sizes of home ranges. The influence of environmental gradients in determining species distributions and community composition has long been recognized by the ecologists. However, the theory and analytical techniques for studying populations in these types of landscapes is still under development. Landscape Structure and Interspecific Interactions Interspecific interactions are important determinants of population dynamics, and landscape structure can influence these interactions. All species interact with predators, parasites, competitors, and so on as the biotic portion of their environment. In breeding birds of forests, nest predation rates are affected by both landscape-level and within-patch habitat patterns (Lloyd et al., 2005). Predation rates increase as patch sizes decline and the landscape-wide abundance of nonforest Author's personal copy Landscape Ecology and Population Dynamics Schematic 499 Forest stands Time since disturbance (years) 0 10 20 40 ≥60 (a) 0 1 2 4 km Stand origin Prior 1880 1900−1910 1936−1950 1986−2005 1880−1899 1911−1935 1951−1985 No data (b) Figure 7 Illustration of a landscape as a shifting mosaic of sites in different stages of recovery from disturbance. In the schematic (a), sites fully recover from disturbance after 60 years of recovery. A landscape in the Pisgah National Forest in North Carolina (b) is a mosaic of forest stands of various ages as measured by time since the year of stand origin. This pattern is a consequence of natural disturbances and human land uses such as timber management. habitats increases. Moreover, landscape structure influences the movement patterns of different predator species so that the risk of encountering a specific predator species depends on the precise location in the landscape (Bergin et al., 2000). Conversely, spatial complexity can promote coexistence between predators and prey. ‘‘Refugia’’ are habitats (or microhabitats) where preys are relatively safe from predators. When predators are abundant, refugia prevent them from driving their prey to extinction. Spatial heterogeneity is one of the principal explanations for the coexistence of competing species (Amarasekare, 2003) in community ecology. Coexistence can be achieved by niche partitioning when species specialize on different habitats or resources. Habitat heterogeneity within and among patches promotes coexistence. Spatial structure and heterogeneity through time are also necessary for the coexistence by the competition–colonization trade-off (Tilman, 1994). In this mechanism, competing species differ in their competitive abilities and ability to colonize new habitat patches. The superior competitors have low colonizing abilities, and the better colonizers are poor competitors. Disturbance periodically kills off local populations of the superior competitor, and for a brief time those sites or patches are empty. The superior colonizer (but poor competitor) quickly occupies the empty sites and holds them until the superior competitor arrives. Regular, localized disturbances permit coexistence by continually providing habitat for the poor competitor (e.g., Debout et al., 2009). Coexistence is possible when the landscape is a shifting mosaic of disturbed patches and sites at different stages of recovery from prior disturbances (Figure 7). Disturbance and its effects on landscape structure through time are a recurring research theme in landscape ecology. Summary Landscape ecology is concerned with the causes and consequences of spatial pattern in nature at multiple spatial scales. Spatial pattern in habitats and resources influences the growth, persistence, and decline of populations. The patch–matrix–corridor model of describing spatial heterogeneity is a useful conceptual model that is directly applicable to populations. Species frequently exhibit a patchy distribution in landscapes as a function of the spatial arrangement of their habitats. This landscape-level pattern (i.e., the landscape structure) determines how populations are organized in Author's personal copy 500 Landscape Ecology and Population Dynamics local breeding groups and how frequently individuals move between these groups. This population structure in turn affects population dynamics. Landscape structure, in combination with life-history traits of a species, determines whether that species exists as a single large population, as a metapopulation of demes connected by movement of individuals, or as a collection of separate isolated populations. An understanding of population structure reveals the conservation challenges facing many species. In landscapes in which habitat is highly fragmented, isolated populations in small patches face a number of problems ranging from loss of genetic diversity to edge effects. Thus, the landscape ecology of populations is related to conservation. Some landscapes exhibit continuous variation in habitat quality, and the patch–matrix–corridor model is not directly applicable. Techniques for describing habitat heterogeneity at multiple scales are available, but understanding how this type of variation affects population dynamics is more challenging. See also: Habitat Loss and Fragmentation. Metapopulations References Allee WC (1931) Animal Aggregations: A Study in General Sociology. Chicago, IL: University of Chicago Press. Allentoft M, Siegismund H, Briggs L, and Andersen L (2009) Microsatellite analysis of the natterjack toad (Bufo calamita) in Denmark: Populations are islands in a fragmented landscape. Conservation Genetics 10: 15–28. Amarasekare P (2003) Competitive coexistence in spatially structured environments: A synthesis. Ecology Letters 6: 1109–1122. Askins RA, Chavez-Ramirez F, Dale BC, et al. (2007) Conservation of grassland birds in North America: Understanding ecological processes in different regions. Auk/Ornithological Monographs 124: 1–46. Baguette M and Van Dyck H (2007) Landscape connectivity and animal behavior: Functional grain as a key determinant for dispersal. Landscape Ecology 22: 1117–1129. Banks-Leite C, Ewers RM, and Metzger J-P (2010) Edge effects as the principal cause of area effects on birds in fragmented secondary forest. Oikos 119: 918–926. Baum KA, Haynes KJ, Dillemuth FP, and Cronin JT (2004) The matrix enhances the effectiveness of corridors and stepping stones. Ecology 85: 2671–2676. Benson TJ, Dinsmore JJ, and Hohman WI (2006) Changes in land cover and breeding bird populations with restoration of riparian habitats in East-central Iowa. Journal of the Iowa Academy of Science 113: 10–16. Bergin TM, Best LB, Freemark KE, and Koehler KJ (2000) Effects of landscape structure on nest predation in roadsides of a midwestern agroecosystem: A multiscale analysis. Landscape Ecology 15: 131–143. Blevins E, Wisely S, and With K (2011) Historical processes and landscape context influence genetic structure in peripheral populations of the collared lizard (Crotaphytus collaris). Landscape Ecology 26: 1125–1136. Caughley G (1994) Directions in conservation biology. Journal of Animal Ecology 63: 215–244. Chamberlain DE and Fuller RJ (2001) Contrasting patterns of change in the distribution and abundance of farmland birds in relation to farming systems in lowland Britain. Global Ecology & Biogeography 10: 399–409. Chardon JP, Adriaensen F, and Matthysen E (2003) Incorporating landscape elements into a connectivity measure: A case study for the Speckled wood butterfly (Pararge aegeria L.). Landscape Ecology 18: 561–573. Collier N, Gardner M, Adams M, McMahon CR, Benkendorff K, and Mackay DA (2010) Contemporary habitat loss reduces genetic diversity in an ecologically specialized butterfly. Journal of Biogeography 37: 1277–1287. Coltman DW, Bowen WD, and Wright JM (1998) Birth weight and neonatal survival of harbour seal pups are positively correlated with genetic variation measured by microsatellites. Proceedings of the Royal Society of London B 265: 803–809. Cushman SA, Evans JS, McGarigal K, and Kiesecker JM (2010) Toward Gleasonian Landscape Ecology: From Communities to Species, From Patches to Pixels. Fort Collins, CO: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station. Research Paper RMRS-RP-84. Debout GDG, Dalecky A, Ngomi AN, and McKey DB (2009) Dynamics of species coexistence: Maintenance of a plant-ant competitive metacommunity. Oikos 118: 873–884. Diamond JM, Terborgh J, Whitcomb RF, et al. (1976) Island biogeography and conservation: Strategy and limitations. Science 193: 1027–1032. Dorner B, Lertzman K, and Fall J (2002) Landscape pattern in topographically complex landscapes: Issues and techniques for analysis. Landscape Ecology 17: 729–743. Driscoll DA (2007) How to find a metapopulation. Canadian Journal of Zoology 85: 1031–1048. Driscoll DA and Weir T (2005) Beetle responses to habitat fragmentation depend on ecological traits, habitat condition, and remnant size. Conservation Biology 19: 182–194. Dunning JB, Danielson BJ, and Pulliam HR (1992) Ecological processes that affect populations in complex landscapes. Oikos 65: 169–175. Ecke F, Lofgren O, and Sorlin D (2002) Population dynamics of small mammals in relation to forest age and structural habitat factors in northern Sweden. Journal of Applied Ecology 39: 781. Fahrig L (2003) Effects of habitat fragmentation on biodiversity. Annual Review of Ecology, Evolution, and Systematics 34: 487–515. Flather CH, Hayward GD, Beissinger SR, and Stephens PA (2011) Minimum viable populations: Is there a ‘‘magic number’’ for conservation practitioners? Trends in Ecology and Evolution 26: 307–316. Flinn KM (2007) Microsite-limited recruitment controls fern colonization of postagricultural forests. Ecology 88: 3103–3114. Flinn KM and Vellend M (2005) Recovery of forest plant communities in postagricultural landscapes. Frontiers in Ecology and the Environment 3: 243–250. Foppen RPB, Chardon JP, and Liefveld W (2000) Understanding the role of sink patches in source-sink metapopulations: Reed warbler in an agricultural landscape. Conservation Biology 14: 1881–1892. Fortin M-J and Dale M (2005) Spatial Analysis: A Guide for Ecologists. New York, NY: Cambridge University Press. Franklin IR (1980) Evolutionary change in small populations. In: Soule ME and Wilcox BA (eds.) An Evolutionary-Ecological Perspective, pp. 135–150. Sunderland, MA: Sinauer. Franklin J (2009) Mapping Species Distributions: Spatial Inference and Prediction. New York: Cambridge University Press. Fraterrigo JM, Turner MG, and Pearson SM (2006) Interactions between past land use, life-history traits and understory spatial heterogeneity. Landscape Ecology 21: 777–790. Gilbert-Norton L, Wilson R, Stevens JR, and Beard KH (2010) A meta-analytic review of corridor effectiveness. Conservation Biology 24: 660–668. Gilpin ME and Soule ME (1986) Minimum viable populations: Processes of species extinction. In: Soule ME (ed.) The Science of Scarcity and Diversity, pp. 19–34. Sunderland, MA: Sinauer Associates. Gonzalez-Megias A, Gomez JM, and Sanchez-Pinero F (2005) Regional dynamics of a patchily distributed herbivore along an altitudinal gradient. Ecological Entomology 30: 706–713. Goodwin BJ (2003) Is landscape connectivity a dependent or independent variable? Landscape Ecology 18: 687–699. Grovenburg TW, Jacques CN, Klaver RW, et al. (2011) Influence of landscape characteristics on migration strategies of white-tailed deer. Journal of Mammalogy 92: 534–543. Gustafson EJ and Parker GR (1994) Using an index of habitat patch proximity for landscape design. Landscape and Urban Planning 29: 117–130. Hackney EE and McGraw JB (2001) Experimental demonstration of an Allee effect in American ginseng. Conservation Biology 15: 129–136. Haddad NM, Hudgens B, Damschen EI, et al. (2011) Assessing positive and negative ecological effects of corridors. In: Liu J, Hull V, Morzillo AT, and Wiens JA (eds.) Sources, Sinks, and Sustainability, pp. 475–503. Cambridge, UK: Cambridge University Press. Haddad NM and Tewksbury JJ (2005) Low-quality habitat corridors as movement conduits for two butterfly species. Ecological Applications 15: 250–257. Hanski I (1999) Metapopulation Ecology. Oxford, UK: Oxford University Press. Hanski I, Moilanen A, and Gyllenberg M (1996a) Minimum viable metapopulation size. The American Naturalist 147: 527–541. Hanski I, Moilanen A, Pakkala T, and Kuussaari M (1996b) The quantitative incidence function model and persistence of an endangered butterfly metapopulation. Conservation Biology 10: 578–590. Author's personal copy Landscape Ecology and Population Dynamics Hanski I and Ovaskainen O (2002) Extinction debt at extinction threshold. Conservation Biology 16: 666–673. Hanski I and Saccheri I (2006) Molecular-level variation affects population growth in a butterfly metapopulation. Public Library of Science Biology 4: e129. Hargis CD, Bissonette JA, and David JL (1998) The behavior of landscape metrics commonly used in the study of habitat fragmentation. Landscape Ecology 13: 167–186. Herkert JR (2007) Evidence for a recent Henslow’s Sparrow population increase in Illinois. Journal of Wildlife Management 71: 1229–1233. Herkert JR, Reinking DL, Wiedenfeld DA, et al. (2003) Effects of prairie fragmentation on the nest success of breeding birds in the Midcontinental United States. Conservation Biology 17: 587–594. Holdo RM, Fryxell JM, Sinclair ARE, Dobson A, and Holt RD (2011) Predicted impact of barriers to migration on the Serengeti wildebeest population. Public Library of Science One 6: 1–7. Honnay O, Jacquemyn H, Bossuyt B, and Hermy M (2005) Forest fragmentation effects on patch occupancy and population viability of herbaceous plant species. New Phytologist 166: 723–736. Jackson HB (2013) Habitat loss and fragmentation. Encyclopedia of Biodiversity, 2nd edn. Burlington, MA: Academic Press. Jewell KJ and Arcese P (2008) Consequences of parasite invasion and land use on the spatial dynamics of host populations. Journal of Applied Ecology 45: 1180–1188. Kindlmann P and Burel F (2008) Connectivity measures: A review. Landscape Ecology 23: 879–890. Krauss J, Bommarco R, Guardiola M, et al. (2010) Habitat fragmentation causes immediate and time-delayed biodiversity loss at different trophic levels. Ecology Letters 13: 597–605. Leidner AK and Haddad NM (2011) Combining measures of dispersal to identify conservation strategies in fragmented landscapes. Conservation Biology 25: 1022–1031. Lewis CW, Hodges KE, Koehler GM, and Mills LS (2011) Influence of stand and landscape features on snowshoe hare abundance in fragmented forests. Journal of Mammalogy 92: 561–567. Lima SL and Dill LM (1990) Behavioral decisions made under the risk of predation: A review and prospectus. Canadian Journal of Zoology 68: 619–640. Lloyd P, Martin TE, Roland LR, Langner U, and Hart MM (2005) Linking demographic effects of habitat fragmentation across landscapes to continental source-sink dynamics. Ecological Applications 15: 1504–1514. Matlack GR (1993) Microenvironment variation within and among deciduous forest edge sites in the eastern United States. Biological Conservation 66: 185–194. Matlack GR (1994a) Plant species migration in a mixed-history forest landscape in eastern North America. Ecology 75: 1491–1502. Matlack GR (1994b) Vegetation dynamics of the forest edge – Trends in space and successional time. The Journal of Ecology 82: 113–123. McCullough DR (ed.) (1996) Metapopulations and Wildlife Conservation. Washington, DC: Island Press. McGarigal K, Tagil S, and Cushman S (2009) Surface metrics: An alternative to patch metrics for the quantification of landscape structure. Landscape Ecology 24: 433–450. Meekins JF and McCarthy BC (2001) Effect of environmental variation on the invasive success of a nonindigenous forest herb. Ecological Applications 11: 1336–1348. Mitrovich MJ, Diffendorfer JE, and Fisher RN (2009) Behavioral response of the coachwhip (Masticophis flagellum) to habitat fragment size and isolation in an urban landscape. Journal of Herpetology 43: 646–656. Palmer MW (1990) Spatial scale and patterns of vegetation, flora and species richness in hardwood forests of the North Carolina Piedmont. Coenoses 5: 89–96. Parker AJ (1982) The topographic relative moisture index: An approach to soilmoisture assessment in mountain terrain. Physical Geography 3: 160–168. Pearson SM (1993) The spatial extent and relative influence of landscape-level factors on wintering bird populations. Landscape Ecology 8: 3–18. Pearson SM and Fraterrigo JM (2011) Habitat quality, niche breath, temporal stochasticity, and the persistence of populations in heterogeneous landscapes. In: Liu J, Hull V, Morzillo AT, and Wiens JA (eds.) Sources, Sinks, and Sustainability, pp. 115–138. Cambridge, UK: Cambridge University Press. Pearson SM, Smith AB, and Turner MG (1998) Forest patch size, land use and mesic forest herbs in the French Broad River Basin, North Carolina. Castanea 63: 382–395. 501 Pearson SM, Turner MG, and Drake JB (1999) Landscape change and habitat availability in the Southern Appalachian Highlands and Olympic Peninsula. Ecological Applications 9: 1288–1304. Peterjohn BG and Sauer JR (1999) Population status of North American grassland birds from the North American Breeding Bird Survey. In: Vickery PD and Herkert JR (eds.) Ecology and Conservation of Grassland Birds Of The Western Hemisphere, pp. 27–44. Studies in Avian Biology No. 19. Pillsbury FC and Miller JR (2008) Habitat and landscape characteristics underlying anuran community structure along an urban-rural gradient. Ecological Applications 18: 1107–1118. Piqueray J, Bisteau E, Cristofoli S, Palm R, Poschlod P, and Mahy G (2011) Plant species extinction debt in a temperate biodiversity hotspot: Community, species and functional traits approaches. Biological Conservation 144: 1619–1629. Price SJ, Marks DR, Howe RW, Hanowski JM, and Niemi GJ (2005) The importance of spatial scale for conservation and assessment of anuran populations in coastal wetlands of the Western Great Lakes, USA. Landscape Ecology 20: 441–454. Pulliam HR (1988) Sources, sinks, and population regulation. American Naturalist 132: 652–661. Radford JQ and Bennett AF (2007) The relative importance of landscape properties for woodland birds in agricultural environments. Journal of Applied Ecology 44: 737–747. Renfrew RB, Ribic CA, Nack JL, and Bollinger EK (2005) Edge avoidance by nesting grassland birds: A futile strategy in a fragmented landscape. Auk 122: 618–636. Ribic CA and Sample DW (2001) Associations of grassland birds with landscape factors in southern Wisconsin. American Midland Naturalist 146: 105. Riiters K, Wickham J, O’Neill R, Jones B, and Smith E (2000) Global-scale patterns of forest fragmentation. Conservation Ecology 4: 3. Riitters KH, O’Neill RV, Wickham JD, and Jones KB (1996) A note on contagion indices for landscape analysis. Landscape Ecology 11: 197–202. Runge Jonathan P, Runge Michael C, and Nichols James D (2006) The role of local populations within a landscape context: Defining and classifying sources and sinks. The American Naturalist 167: 925–938. Saccheri I, Kuussaari M, Kankare M, Vikman P, Fortelius W, and Hanski I (1998) Inbreeding and extinction in a butterfly metapopulation. Nature 392: 491. Saunders DA, Hobbs RJ, and Margules CR (1991) Biological consequences of ecosystem fragmentation: A review. Conservation Biology 5: 18–32. Seydack AHW, Vermeulen C, and Huisamen J (2000) Habitat quality and the decline of an African elephant population: Implications for conservation. South African Journal of Wildlife Research 30: 34. Shaffer ML (1981) Minimum population sizes for species conservation. BioScience 31: 131–134. Shen G, Feng C, Xie Z, Ouyang Z, Li J, and Pascal M (2008) Proposed conservation landscape for giant pandas in the Minshan Mountains, China. Conservation Biology 22: 1144–1153. Simon SA, Collins TK, Kauffman GL, McNab WH, and Ulrey CJ (2005) Ecological zones in the Southern Appalachians: First Approximation. Asheville, NC: US Department of Agriculture, Southern Research Station. Forest Service Research Paper SRS-41. Slatkin M (1985) Gene flow in natural populations. Annual Review of Ecology & Systematics 16: 393–430. Snep R and Ottburg F (2008) The ’habitat backbone’ as strategy to conserve pioneer species in dynamic port habitats: Lessons from the natterjack toad (Bufo calamita) in the Port of Antwerp (Belgium). Landscape Ecology 23: 1277–1289. Steffan-Dewenter I, Munzenberg U, and Tscharntke T (2001) Pollination, seed set and seed predation on a landscape scale. Proceedings of the Royal Society of London: Biological Sciences 268: 1685–1690. Stephens PA, Sutherland WJ, and Freckleton RP (1999) What Is the Allee Effect? Oikos 87: 185–190. Sutton FM and Morgan JW (2009) Functional traits and prior abundance explain native plant extirpation in a fragmented woodland landscape. Journal of Ecology 97: 718–727. Thompson LM, van Manen FT, Schlarbaum SE, and DePoy M (2006) A spatial modeling approach to identify potential butternut restoration sites in Mammoth Cave National Park. Restoration Ecology 14: 289–296. Tilman D (1994) Competition and biodiversity in spatially structured habitats. Ecology 75: 2–16. Traill LW, Brook BW, Frankham RR, and Bradshaw CJA (2010) Pragmatic population viability targets in a rapidly changing world. Biological Conservation 143: 28–34. Turner SJ, O’Neill. RV, Conley W, Conley MR, and Humphries HC (1991) Pattern and scale: Statistics for landscape ecology. In: Turner MG and Gardner RH Author's personal copy 502 Landscape Ecology and Population Dynamics (eds.) Quantitative Methods in Landscape Ecology: The Analysis and Interpretation of Landscape Heterogeneity, pp. 17–49. New York: SpringerVerlag. Veech JA (2006) A comparison of landscapes occupied by increasing and decreasing populations of grassland birds. Conservation Biology 20: 1422–1432. Vellend M (2003) Habitat loss inhibits recovery of plant diversity as forests regrow. Ecology 84: 1158–1164. Vellend M, Verheyen K, Flinn KM, et al. (2007) Homogenization of forest plant communities and weakening of species-environment relationships via agricultural land use. Journal of Ecology 95: 565–573. Vergara P (2011) Matrix-dependent corridor effectiveness and the abundance of forest birds in fragmented landscapes. Landscape Ecology 26: 1085–1096. Verheyen K, Guntenspergen GR, Biesbrouck B, and Hermy M (2003) An integrated analysis of the effects of past land use on forest herb colonization at the landscape scale. Journal of Ecology 91: 731–742. Wang Y-H, Yang K-C, Bridgman C, and Lin L-K (2008) Habitat suitability modelling to correlate gene flow with landscape connectivity. Landscape Ecology 23: 989–1000. Wiens JA and Milne BT (1989) Scaling of ‘‘landscapes’’ in landscape ecology, or, landscape ecology from a beetle’s perspective. Landscape Ecology 3: 87–96. Wiens JA, Stenseth NC, Horne BV, and Ims RA (1993) Ecological Mechanisms and Landscape Ecology. Oikos 66: 369–380.