《森林生态学》课程教学资源（生态学热点）Issues in ecology - 16 景观连通性在规划和实施保护中的作用 The Role of Landscape Connectivity in Planning and Implementing Conservation and Restoration Priorities

ISSUES IN ECOLOGY NUMBER SIXTEEN FALL 2012 Box 1.Habitat Fragmentation and Increased Disease Transmissivity a to be m for ulatio land bacte that c rum (Bon urg ely resul bined with release from natural e found ins ts.pthe nay those near larger forest fragments Area (ha) gm b)Pe apugceieshe Source: risk. ecteodwt pathogens and parasites for which the host d,and their porential constitute the true x D).Ho "across which both fragment the ter estrial environment and host subpopulations.which then may bec ome en aquatic syst habitats,they can simultaneous e as cor scape fragme tation ma duits for some invasive species,such as cheat from lisease.For example mia pestis ow star thistl n orado prairie dog (C that benefit from the onenin. more closely grouped together. ion and Measuring,Analyzing and increased host connectivity,it does not neces. Designing Landscape Connectivity ect Mea host abundance and alter host distributior and these changes can increase connectivity for and policy makers,as Geographic Information The Ecological Society of America.esahg@esa.org esa 5

© The Ecological Society of America • esahq@esa.org esa 5 ISSUES IN ECOLOGY NUMBER SIXTEEN FALL 2012 molitrix), and bighead carp (H. nobilis)) in the Mississippi River watershed, and their potential spread into the Great Lakes, illustrate how human-constructed connections (canals) can both fragment the terrestrial environment and provide new corridors between aquatic systems. Similarly, while roads can fragment vegetated habitats, they can simultaneously serve as con￾duits for some invasive species, such as cheat￾grass (Bromus tectorum), yellow star thistle (Centaurea solstitialis), and other invasive species that benefit from the openings created by roads. Connectivity for pathogens and parasites is largely a function of host distribution and abun￾dance. While disease persistence benefits from increased host connectivity, it does not neces￾sarily follow that these conditions are optimized only in well-connected landscapes. Landscape disturbance and fragmentation can increase host abundance and alter host distribution, and these changes can increase connectivity for pathogens and parasites for which the hosts constitute the true “landscape” across which movement occurs (Box 1). However, fragmen￾tation may also lead to the isolation of smaller host subpopulations, which then may become more susceptible to disease or invasions. In other situations, isolation resulting from land￾scape fragmentation may protect a population from disease. For example, plague (Yersinia pestis) in Colorado prairie dog (Cynomys ludovicianus) populations was shown to be less prevalent in more remote, isolated populations than in those more closely grouped together. Measuring, Analyzing and Designing Landscape Connectivity Measuring structural connectivity has increas￾ingly become a routine objective of researchers and policy makers, as Geographic Information Box 1. Habitat Fragmentation and Increased Disease Transmissivity An important consequence of fragmentation in forested habitats is the loss of species diversity. Those species that thrive in fragmented habitats tend to be more generalist or opportunistic, or have traits such as smaller home range require￾ments and tolerance for higher densities. Fragmentation can actually increase connectiv￾ity from the perspective of a disease-causing pathogen. Higher densities of hosts increase opportunities for transmissivity, and the host population is the true “landscape” across which pathogen movement occurs. This is the case for the tick-transmitted bacterium (Borrelia burgdor￾feri) that causes Lyme disease. Its host, the white-footed mouse (Peromyscus leucopus), has become increasingly common in small forest fragments (<2 ha) in New England, likely result￾ing from its small home range requirements combined with release from competitors and predators in smaller forest patches. P. leucopus is the principal natural reservoir for Lyme dis￾ease. Higher densities of ticks infested with B. burgdorferi are found in smaller forest fragments (Figure 1), which may result from higher densities of white-footed mouse in these smaller frag￾ments, presenting more opportunities for ticks to feed on the mice. Consequently, humans living near these small forest fragments may have a higher risk of exposure to Lyme disease relative to those near larger forest fragments. Figure 1. Relationship between measures of Lyme disease risk and forest patch area in a frag￾mented landscape in New York state. a) Density of nymphal ticks is higher in smaller forest fragments. b) Percentage of nymphal ticks infected with the bacterium Borrelia burgdorferi is higher in smaller forest fragments. c) Density of nymphal ticks infected with the B. burgdorferi is higher in smaller forest fragments. (Source: Allan, B.F., F. Keesing, and R.S. Ostfeld. 2003. Effect of forest fragmentation on Lyme disease risk. Conservation Biology. 17: 267–272). Image used with permission of John Wiley and Sons. (a) (b) (c)

ISSUES IN ECOLOGY NUMBER SIXTEEN FALL 2012 System(GIS)and remote sensing tools needs.Thus,recent developments in connec affordable,and ne a s isms can be logistically complicated. with a functional approach that highlights scan rac only rela t requirements controlled experiments addressing movements Modeling Approaches for dispersal at relevant Quantifying may more accurately and efficiently reflect across large emented in a GlS env s of tracking individual animals and inte grate only those mo vements that produce approach has specific data meaning popu tion impacts ispe require input from b s to help C. comine of this anpr isthat current geneti ns may not reflect the impact of current ally for species wit comes. Least-cost analysis identifics the least human perse by past epidemics by oal adaptation.which can drive etic where ost"may reflect the actual energ 0ecradland expended to move over the area,mortality risk cting or impact on A common product of connectiv ity analysis e of habitat.Habitats that the anima hil n D nThe le tion of cells that has the lowest cumulative unique strengths as the path as a pa w popu sistance (an indica of how well a land patch)to the other endpoint. cape can be traversed by a giv species),a can 1 of information about s hahitat nrefer can be lysts lon methods to rig sly n in red in the r anel o -specific resistar om at e which is a swath of cells expected to provide a ow,genetic at use, ow-co r movemen ance.hased on the extent to which land patchesa esults in higher costs.This latte cover,in hay be y ar mpacts measures may be useful for some gen cale of perc ption andh may not be able to cor sider total 1.C ing species-specific movements and so referred to as 6 esa The Ecological Society of America.esahg@esa ora

ISSUES IN ECOLOGY NUMBER SIXTEEN FALL 2012 6 esa © The Ecological Society of America • esahq@esa.org System (GIS) and remote sensing tools become more widely available, affordable, and scalable. However, measuring functional con￾nectivity using the movements of individual organisms can be logistically complicated. Even the largest studies using the most appro￾priate technologies can track only relatively few individuals over modest time periods, and controlled experiments addressing movements and dispersal at relevant scales are extremely difficult to implement. One way to address this difficulty is to measure gene flow, which may more accurately and efficiently reflect functional connectivity across large landscapes. Genetic studies avoid the logistic and financial costs of tracking individual animals and inte￾grate only those movements that produce meaningful population impacts – dispersals that result in breeding or emigration. A short￾coming of this approach is that current genetic patterns may not reflect the impact of current landscape features, especially for species with large population sizes or long generation times, or species affected by unobserved events, such as genetic bottlenecks caused by past epidemics or human persecution. In addition, genetic connectivity may be masked in some instances by local adaptation, which can drive genetic distinctiveness even in a well-connected land￾scape, by selecting for particular characteristics of the local environment. A common product of connectivity analysis is a map of predicted core areas, linkage zones, or barriers. Such maps often become the basis for management actions. Several tools can be used to map these features, and each has unique strengths and weaknesses. All of the approaches described in the next section depend on accurately defining landscape resistance (an indication of how well a land￾scape can be traversed by a given species), a challenging task when only a limited amount of information about species habitat prefer￾ences is available. Furthermore, connectivity models can be difficult to validate. Several research teams are working to develop methods to rigorously estimate species-specific resistance from data on gene flow, genetic distances, habitat use, and movement paths. Simple estimates of resis￾tance, based on the extent to which land￾scapes are impacted by roads, loss of natural land cover, increased edge effects, spread of invasive species, and other direct human impacts measures may be useful for some gen￾eralist species, but are insufficient for address￾ing species-specific movements and habitat needs. Thus, recent developments in connec￾tivity modeling combine a structural land￾scape approach, identifying both the potential for and obstacles to long-term habitat shifts, with a functional approach that highlights the specific connectivity needs of species with restricted habitat requirements. Modeling Approaches for Identifying and Quantifying Landscape Connectivity We describe five widely-used analytical approaches, all implemented in a GIS envi￾ronment, to assist planners in mapping and prioritizing landscape connections. Each approach has specific data requirements that often require input from biologists to help define model parameters. In addition, each approach is designed to meet different objec￾tives and will, therefore, produce different out￾comes. Least-cost analysis identifies the least costly route that an animal can take from one area to another. The method assumes that the animal incurs a cost as it moves over an area, where “cost” may reflect the actual energy expended to move over the area, mortality risk, or impact on future reproductive potential. In practice, cost is usually estimated simply as the inverse of habitat suitability. Habitats that the animal favors are assigned low cost while unsuitable habitats are assigned high cost. The least-cost path is the contiguous collec￾tion of cells that has the lowest cumulative cost as the path crosses from one endpoint (such as a park, natural area, or known popu￾lation; sometimes referred to as a node or patch) to the other endpoint. Computers using GIS software can easily identify this path. Because the least-cost path is only one cell wide (for example, the center panel in Figure 2), it is often not a realistic area to pro￾pose for conservation. Therefore, analysts usu￾ally identify the least-cost corridor (shown in red in the panel on the right in Figure 2), which is a swath of cells expected to provide a low-cost route for movement. Increased distance between two nodes or patches also results in higher costs. This latter assumption is important, in that some species may be able to identify and take advantage of shorter linkages, while others operate at a finer scale of perception and therefore may not be able to consider total corridor length. Correctly assigning these cost values (also referred to as