Top
divider

Wildlife Habitat Linkages in the Eastern Adirondacks: Applying Functional Connectivity Modeling to Conservation Planning for Three Focal Species

Author: Rose A. Graves and Deane Wang Back to Table of Contents >> From Peer Review - Volume 18 (2012) (2012)

As habitat loss and fragmentation increase across the northeastern United States, identifying and prioritizing connecting routes between protected areas has taken on new urgency.  Protecting habitat linkages, or corridors, in which species can live and move between core habitats is a useful strategy for maintaining biodiversity, reducing the negative effects of habitat fragmentation, and potentially mitigating effects of climate change.  Spatial models are an informative tool to predict the best locations for conservation corridors by incorporating specific landscape features and the available information on wildlife behavior and preferences.  As large landscape conservation initiatives gain traction in the conservation community, conservation planners can use spatial tools to conduct connectivity analyses as opposed to creating conservation plans through ad-hoc methods.

Here we present a case study using CorridorDesigner, a free software program, and modeled landscape resistance surfaces based on expert knowledge to predict a habitat linkage location that could provide functional connectivity for three focal species: black bear (Ursus americanus), bobcat (Lynx rufus), and fisher (Martes pennanti) in the Split Rock Wildway conservation planning area (SRW) in Essex County, NY.  The analysis area was limited to the SRW in order to provide comparison with an existing conservation effort.  The methods described in this paper provide a cost-effective, science-based, and transparent way to assess habitat connectivity for conservation planning.  This assessment: 1) provides a functional habitat linkage for three mammal species, 2) evaluates the uncertainty in resistance surfaces used to predict that habitat linkage, and 3) compares the predicted functional habitat linkage to an existing ad-hoc linkage effort in the same conservation planning area.  In the SRW, our model suggests that the best functional habitat linkage for black bear, bobcat, and fisher is located south of a current ad-hoc initiative.  This functional habitat linkage location differed significantly from the ad-hoc linkage in location as well as in the perceived resistance to movement for each species.  Multiple model simulations tended to converge on the same functional habitat linkage.  Our results suggest that the predicted functional habitat linkage should be included in conservation plans aimed at maintaining landscape connections between the Split Rock Wild Forest and the larger wild areas of the Adirondack Park. 

 Introduction

Over the last two decades, ecologists have recognized the importance of landscape connectivity as it relates to the conservation of biodiversity, mitigation of habitat fragmentation, and the potential for species to respond to changes in land use and climate (Bennett, 2003; Crooks & Sanjayan, 2006a; Ewers & Didham, 2006; Giles, 1998; Lindenmayer et al., 2008; Soule and Noss, 1998).  Increased landscape connectivity can help prevent negative population-level effects due to isolation and could facilitate increased population sizes, viability, and dispersal of species (Crooks & Sanjayan, 2006b; Hanski, 1999; Harrison, 1992; Noss, 1983; Noss, 1987).  It is generally assumed that connected reserves support larger populations that are less likely to succumb to local random extinction and that landscape connectivity facilitates movement of organisms among patches resulting in greater genetic exchange between populations and minimizing the loss of genetic variation.  Conservation of habitat linkages or movement corridors has been recognized as one potential way to maintain habitat connectivity (Beier & Noss, 1998; Crooks & Sanjayan 2006a; Dixon et al., 2006; Driezen et al., 2007; Haddad et al., 2000; Huck et al., 2011).  The connectivity of a landscape for wildlife can be described as: functional, referring to how a particular landscape’s structure affects an animal’s behavior; or structural, which refers to the spatial arrangement and proximity of habitat patches, regardless of an individual animal’s perception.  While many conservation scientists agree that connected landscapes are desirable, there is little agreement on how to measure and incorporate landscape connectivity into conservation plans (Berke, 2007; Kindlmann & Burel, 2008).

At the same time that landscape-level conservation strategies are increasingly encouraged (Bateson, 2005; Crooks & Sanjayan, 2006a; Hilty et al., 2006; Noss, 1987; Prato & Fagre, 2007), it has been recognized that systematic conservation planning is more effective than ad-hoc planning of the past (Chetkiewicz et al., 2006; Margules & Pressey, 2000; Noss, 2003).  However, it is often difficult for small land conservation organizations to incorporate these ideas into their local work.  Small land organizations often rely on connectivity analyses conducted by larger organizations, which may be inappropriate in scale and extent to allow for local identification and prioritization of land conservation projects (Berke, 2007; Groves et al., 2002).  Conservation initiatives can be more successful by incorporating an understanding of animal movement across a landscape, but availability of empirical data can be a challenge for organizations with limited financial and human resources.

Connectivity conservation in the Northern Appalachian/Acadian ecoregion provides one example of this challenge.  An ecoregional analysis conducted by Canadian and U.S. scientists working with the organization Two Countries, One Forest (2C1F) identified five top conservation priorities to maintain the ecological health and connectivity across the ecoregion (Bateson, 2005; Trombulak et al., 2008).  They concluded that the link between the Adirondack Park and Vermont’s Green Mountains was vital to the health of the Northern Appalachian/Acadian ecoregion and was likely to be compromised in the future; however, they did not provide specific actionable conservation priorities within that linkage.  More recent analysis provided several possible linkages between the Adirondack and Green Mountains but remained focused on a regional scale (Baldwin et al., 2010).

The Split Rock Wildway Planning Area (SRW) located in Essex County, NY and within an area identified by 2C1F scientists as highly irreplaceable and highly threatened represents an opportunity to maintain an important stepping stone within this regional linkage area (Figure 1).  Several local partners and conservation groups identified this area as a priority in which to restore and preserve habitat with the goal of linking wilderness areas in the northern Adirondack Park with the Champlain Lowlands and Lake Champlain to the east (Northeast Wilderness Trust, 2009).  No systematic assessment of landscape connectivity in the SRW has been conducted and GIS data identifying conservation priorities are at a scale of 10 km2 or higher (Baldwin et al., 2010; Trombulak et al., 2008).  Previous conservation planning within the SRW has been ad-hoc, has focused on structural connectivity to link previously conserved lands, and included the identification of an ad-hoc habitat linkage based on local conservationists’ input (New York State Department of Environmental Conservation, 2005).  

Figure 1.  The Split Rock Wildway Planning Area in Essex County, NY located at the eastern edge of the Adirondack Park.  Inset shows Two Countries, One Forest in the northeastern United States and Canada, with arrows indicating wildlife corridor conservation priority locations (Trombulak et al., 2008).

Figure 1. The Split Rock Wildway Planning Area in Essex County, NY located at the eastern edge of the Adirondack Park. Inset shows Two Countries, One Forest in the northeastern United States and Canada, with arrows indicating wildlife corridor conservation priority locations (Trombulak et al., 2008).

Geographic information system (GIS) and cost-distance models provide an opportunity to use species-specific information regarding habitat preferences “to identify critical corridors and permeable habitats” and to assess the functional connectivity of a landscape (Walker & Craighead, 1997). Cost-distance models, which represent landscape resistance by assigning an ecological cost to an animal crossing a particular landscape, have been used to describe connectivity for amphibians, birds, and mammals (see Sawyer et al., 2011).  Many conservation organizations and government agencies have also used cost-distance modeling to guide conservation planning (Bates & Jones, 2007; Brown et al., 2010; Long, 2007; Singleton et al., 2002; Singleton & Lehmkuhl, 2001; Zeh & Marangolo, 2010).  The difficulty and the financial and time costs associated with collecting empirical data on animal movement and genetic flow across a landscape make GIS models attractive to organizations interested in understanding functional connectivity in their conservation priority areas and several free software tools exist to help organizations do this (Calabrese & Fagan, 2004; Carroll, 2010; Majka et al., 2007; Shah & McRae, 2008; Theobald et al., 2006).  Where detailed species and movement data are lacking, models based on expert knowledge, using peer-reviewed literature and personal communication with wildlife biologists, have been used in conservation designs and connectivity analyses (Beier et al., 2008; Chetkiewicz et al., 2006; Sawyer et al., 2011). 

We used CorridorDesigner, a free GIS toolkit, and species-specific GIS models of landscape resistance to identify a functional habitat linkage within the SRW for three focal species: black bear (Ursus americanus), bobcat (Lynx rufus), and fisher (Martes pennanti).  We chose these focal species based on the availability of literature on movement patterns and habitat use, their status as wide-ranging vertebrates with large area requirements, and their inclusion in existing conservation plans in the region (Brown et al., 2010; Long, 2007; Zeh & Marangolo, 2010).  Black bear are wide-ranging mammals and are considered landscape species that require a variety of habitats to meet their life history requirements (Costello, 1992; Rogers & Allen, 1987; Schoen, 1990; Simek, 1995).  Their seasonal movements, large home range sizes, and relatively low natural densities make black bears vulnerable to habitat fragmentation (Hammond, 2002; Lariviere, 2001).  Bobcats are also described as sensitive to habitat fragmentation and urbanization (Crooks, 2002; Hansen, 2007; Riley et al., 2003).  Connectivity of habitats may be particularly important for dispersing juvenile bobcats (Johnson et al., 2010).  Fisher populations have been described as inherently unstable and subject to local extinctions and colonizations (Buskirk & Powell, 1994; Powell et al., 2003).  While fisher are broadly described as forest carnivores sensitive to human development and disturbance, fishers fitted with global positioning system (GPS) collars in urban Albany, NY adjusted daily activity patterns but continued to select for connected forest habitat (LaPoint & Kays, 2011). 

We sought to answer the following questions: 1) What habitat linkage provides the best functional connectivity for all three focal species within the SRW? 2) How does uncertainty in the model parameter values affect the landscape resistance models used to predict the habitat linkage? and 3) How does the predicted functional habitat linkage compare to an existing ad-hoc linkage effort in the same conservation planning area?  Finally, we hope to illustrate a user-friendly method for assessing connectivity at a local scale within the budget of a small non-profit conservation organization.

Study Area

The study area encompasses approximately 900 km2 of primarily forested and agricultural land in the western Lake Champlain Basin (Figure 1).  Fully within the boundary of the Adirondack Park, it is a mosaic of public and private ownership containing a diversity of habitats ranging in elevation from 29 m at the Lake shore to 1,219 m at the top of Giant Mountain.  Forests within the study area include spruce-fir (Picea spp., Abies balsamea), evergreen-northern hardwood, and mesic upland hardwoods dominated by sugar maple (Acer saccharum), American beech (Fagus grandifolia), and yellow birch (Betula allegheniensis) as well as patches of oak (Quercus spp.) forest and successional hardwoods.  Non-forest communities include forested and open wetlands, open water, current and abandoned agricultural lands, and low- to medium- intensity development areas.

 A 139-km2 wildland block owned by the State of New York, including Jay and Giant Wilderness Areas and Hurricane Mountain Primitive Use Area, is located on the western side of the study area.  The eastern side of the study area borders Lake Champlain and includes a second New York State-owned wildland block- Split Rock Wild Forest.  The 15-km2 Split Rock Wild Forest represents the largest block of undeveloped land on the western shore of Lake Champlain.  The geodesic (non-Euclidean) distance between the Jay and Giant Wilderness Areas block and Split Rock Wild Forest is 21 km, and this area, referred to as the Split Rock Wildway conservation planning area, is intersected by one major interstate, three major state routes, and a multiple local roads.  The identified ad-hoc habitat linkage follows a route southeast from the Jay Wilderness Area to Split Rock Wild Forest (Northeast Wilderness Trust, 2009) (Figure 2).   

Figure 2.  The ad-hoc habitat linkage plan follows a route from the Jay Wilderness Area southeast to Split Rock Wild Forest, crossing forests, wetlands, and agricultural lands (Northeast Wilderness Trust, 2009).

Figure 2. The ad-hoc habitat linkage plan follows a route from the Jay Wilderness Area southeast to Split Rock Wild Forest, crossing forests, wetlands, and agricultural lands (Northeast Wilderness Trust, 2009).

 

Methods

Many approaches exist for modeling and identifying habitat linkages (Hargrove et al., 2004; Majka et al., 2007; Shah & McRae, 2008; Theobald et al., 2006; Urban & Keitt, 2001; Walker & Craighead, 1997). We chose to use CorridorDesigner to conduct cost-distance modeling to identify habitat linkages between habitat blocks (Adriaensen et al., 2003; Beier et al., 2008; Beier et al., 2006; Majka et al., 2007; Singleton et al., 2002).  Cost-distance methods use GIS models of a species’ perception of the landscape based on the known or assumed ability of that species to move across different landscape features (i.e., the permeability or ecological cost of those features) (Adriaensen et al., 2003; Kirk & Zielinski, 2010; Singleton & Lehmkuhl, 2001; Walker et al., 2007).  Algorithms within CorridorDesigner calculate the minimum cost-distance paths across a modeled landscape wherein the cumulative ecological cost to move between two points is minimized (Majka et al., 2007).  Identifying habitat linkages first required creating GIS models of landscape resistance for each of the focal species which were used in CorridorDesigner to calculate minimum cost-distance habitat linkages.

Modeling Landscape Resistance

The process for developing a GIS raster (pixel) resistance surface followed three steps.  First, we identified which landscape factors to include in the model and the factor weights in determining overall resistance.  We then scored the classes within each landscape factor according to their permeability (e.g., a high permeability has a low resistance to movement).  Third, we combined landscape factors using GIS to assign a score to each pixel (Beier et al., 2008; Beier et al., 2006; Sawyer et al., 2011).  The first two steps of our landscape resistance models relied on expert-knowledge to assign parameters (Clevenger et al., 2002; Beier et al., 2008).

For each focal species, we reviewed the scientific literature associated with habitat requirements and movement patterns to develop species- and study area-specific models of resistance.  Where possible we used studies located in the northeastern United States, paying particular attention to studies of the focal species in the Adirondack Park of New York, and in the Champlain Valley and Green Mountains of Vermont.  From this review, we selected landscape factors that are most likely to influence movement of the focal species.  Because of the prevalence of literature describing habitat use and the corresponding lack of literature related directly to animal movement or dispersal, we followed the major assumption that animals choose travel routes based on the same factors they use to select habitat (Beier et al., 2008; Chetkiewicz et al., 2006; Sawyer et al., 2011).  The most important landscape factors were not the same for all species.  The final landscape factors used in the model were those identified in the literature review as most important for determining species-specific landscape resistance and also available or derivable as GIS data.  Some factors that likely influence species movements across the landscape (e.g., density of conspecific competitors or prey density) were not available in GIS data layers and thus not included in the model.  The final selection of landscape factors was species-specific and was reviewed by biologists who studied the focal species.

For each species resistance model, the landscape factors were assigned a weight based on the relative importance of each factor in determining the resistance of any particular landscape pixel for the focal species (Table 1).  Weights for a species summed to 100.  Within each landscape factor included in a species model, we assigned a permeability score to each class (e.g., deciduous forest or high intensity development classes within the land cover landscape factor).  Permeability scores were set, based on our best estimates from the literature review and input by biologists, such that 0 indicated a completely resistant (non-traversable), unusable habitat and 100 represented optimal habitat.  Scores were assigned as multiples of 10 with 10-30 representing strongly avoided habitat, 40-60 representing marginal habitat, 60-80 as suboptimal habitat, and 80-100 as strongly preferred habitat/completely traversable (Beier et al., 2008; Beier et al., 2009; Majka et al., 2007).  After the factor weight and resistance scores were assigned for each landscape factor, we assigned a minimum and maximum possible value to provide an assumed biologically plausible range of uncertainty for each model parameter (Table 2) (Johnson & Gillingham, 2004).  We based the final assignment of permeability scores on the available literature, existing habitat and connectivity models, and the opinions of species experts who reviewed our habitat suitability scores (Krohn, 2010; Marangelo, 2010; Morse, 2010; Reed, 2010). 

Land cover was considered the most important landscape factor for all three species and received the highest weight in the models. Straight-line (Euclidean) distance to roads was included in all three resistance models. Other landscape factors included Euclidean distance to streams (bobcat and fisher), topographic position (black bear), percent canopy cover (fisher), and core/edge habitat (bobcat) (see Appendices A1, A2, and A3).

Black Bear: Based on the literature and expert opinion, black bear respond primarily to forest cover and human development, as well as the availability of foraging habitat (Table 1).  Movements tend to be associated with food availability and with avoidance of certain features that can be captured in land cover data.  Hammond (2002) suggested that black bears in Vermont avoid areas up to 200 m from roads during all seasons.  Other studies have found road avoidance for black bears to range from 100 m to 1,000 m (Brandenburg, 1996; Brody & Pelton, 1989; Clark et al., 1993; Fecske et al., 2002; Kasworm & Manley, 1990; Lentz et al., 1980; Villarrubia, 1982).  Roads contribute to mortality of black bears and were found to be the leading cause of non-hunter mortality in the central Adirondacks (Simek, 1995), thus distance to roads was included as a factor in the black bear landscape resistance model.  Topographic position was also included in the model as black bears in regions similar to the study area may preferentially use ridgelines as travel corridors (Reed, 2010).

Bobcat: Land cover was identified as the most important factor influencing bobcat movement (Table 1). In addition, bobcats appear to avoid roads (Abouelezz, 2009; Long, 2006; Lovallo & Anderson, 1996b). Bobcats in the upper Champlain Valley and Green Mountains of Vermont tended to avoid “deep forest” (>300 m from edge) and used forest and wetland edge (0 – 30 m from edge) disproportionately, suggesting that forest edges, wetlands, and streams are important movement corridors to connect larger blocks of habitat (Abouelezz, 2009).  We included a core/edge habitat landscape factor as well as a distance to stream factor in the bobcat landscape resistance model.

Fisher: Habitat preferences for fisher vary regionally but can be generalized by habitat type; therefore, land cover was assigned as the most important landscape factor for fisher (Table 1).  Allen (1983) suggested that optimal habitat was based on four variables: canopy closure, stand diversity, average size of stand, and percent deciduous cover.  Based on the importance ascribed to canopy closure and fishers’ often-cited avoidance of open areas, we chose to include a canopy cover landscape factor in the model.  Fishers will travel through forested patches of non-preferred habitat to contiguous patches of preferred habitat and use riparian habitats as important travel corridors (Aubry & Houston, 1992; Buck et al., 1994; Heinemeyer, 1993; Jones & Garton, 1994).  Thus we included distance to streams as an important landscape factor.  Like black bear and bobcat, evidence suggests fisher avoid roads and roads in our study area contribute largely to the mortality of fisher (Coulter, 1966; Fisher, 2004; LaPoint, 2007).

For each species, GIS data layers were obtained for each landscape factor and transformed to Albers equal-area conic projection, 30 m x 30 m resolution raster layers using ArcMap 9.3.2.  Using the CorridorDesigner toolbox for ArcGIS (Majka et al., 2007), we combined the landscape factors using a weighted geometric mean, as recommended by Beier et al. (2008), to produce an overall resistance score between 0 and 100 for each pixel (Majka et al., 2007).

Table 1.  Summary of habitat factor weights and data sources used in the habitat suitability model as well as the identified biologically plausible range for each weight.  This range provided the basis for analyzing the uncertainty associated with the identified habitat linkages.

Table 1. Summary of habitat factor weights and data sources used in the habitat suitability model as well as the identified biologically plausible range for each weight. This range provided the basis for analyzing the uncertainty associated with the identified habitat linkages.

Identifying Habitat Linkages

We used ArcGIS 9.3.2 and the CorridorDesigner toolbox (Majka et al. 2007) to identify habitat linkages in the SRW.  CorridorDesigner allows planners to calculate the conditional minimum transit cost (CMTC), an extension of the minimum cost-distance, between two patches.  CorridorDesigner requires inputs of landscape resistance layers and designated starting and ending patches (vector data layers).  The tool randomly assigns multiple points within the starting and ending patches and calculates the pathways with the lowest cumulative resistance, or highest permeability, for each species (Majka et al., 2007; Pinto & Keitt, 2009; Sawyer et al., 2011).

We defined the starting and ending patches as the Jay and Giant Wilderness Areas block and the Split Rock Wild Forest, respectively.  We constrained the program such that CorridorDesigner only chose starting points with a neighborhood (10 km2) average suitability of 60 or greater.  These criteria were chosen to ensure that sources within the starting and ending patches were large enough and of suitable habitat to support the focal species.  We ran the CorridorDesigner toolkit for each species, setting a moving window of analysis of 200 m for black bear and 100 m for both bobcat and fisher.  We required that the final path for each species meet the criteria of having an average width of over 1 km and a minimum width of 300 m (Beier et al., 2008; Majka et al., 2007).  Finally, we merged the individual focal species paths to create a single functional habitat linkage for all three species in the SRW.

Alternate Scenarios

A major criticism of this habitat linkage modeling procedure involves the uncertainty surrounding landscape factor scores and weights (Beier et al., 2009; Clevenger et al., 2002; Sawyer et al., 2011).  To examine the sensitivity of the resistance models (on which the habitat linkages are based) to changes in landscape factor weights and scores, we varied parameter inputs to create alternate scenarios for each focal species (i.e., black bear, 10; bobcat, 12; fisher, 12) following a methodology similar to Beier et al. (2009).

We varied factor weights and class permeability scores based on the biologically plausible range identified for each species (Table 1 and Appendices A1, A2, and A3).  For each landscape factor, we determined the mean permeability score.  We then created a compressed scenario for each landscape factor by reassigning the permeability scores to the value closest to the mean score while remaining in the biologically plausible range for each factor (Figure 5).  Similarly, a dispersed scenario was created for each landscape factor where the permeability scores were set to the value within the biologically plausible range, which maximized the difference from the mean score.  We varied each factor’s permeability scores while keeping the other factors constant and created two scenarios where we altered all of the landscape factors at once either compressing or dispersing all scores (Appendices A1, A2, and A3).  Finally, we created two alternate scenarios for each species wherein the weights of the landscape factors were either compressed or dispersed.  For each scenario, we followed the procedure described above to create GIS raster layers of landscape resistance.  The resistance surfaces calculated from these alternate scenarios were the same extent and spatial scale as the original resistance surfaces.

Each model run produced a map containing one observation per 30 m x 30 m pixel with a unique score, resulting in extremely large datasets (n > 990,000).  To assess how each change in the parameter values affected the resulting resistance surfaces, we performed pixel-by-pixel comparisons using a Kappa test which provided an index that accounted for agreement due simply to chance between the alternate resistance surface models and the original resistance surface model (King, 2004; Levine et al., 2009; Monserud & Leemans, 1992).  Generally, Kappa values closer to 1.0 indicate high agreement between two scenarios with 0.80 – 1.0 indicating almost perfect agreement and 0.01- 0.20 indicating only slight agreement (Viera & Garrett, 2005).  For each species, we also calculated the percent overlap between the alternate scenario habitat linkages and the predicted functional habitat linkage using spatial analysis tools within ArcGIS.

Comparison of Functional and Ad-hoc Habitat Linkages

To assess differences between the functional habitat linkage and a previously identified ad-hoc habitat linkage (Northeast Wilderness Trust, 2009), we compared several landscape characteristics including the habitat quality for each species as described by the modeled landscape resistance surface (Beier et al., 2008).  We assumed that areas modeled as low resistance corresponded to high habitat quality and vice versa (Beier et al., 2008; Beier et al., 2009; Majka et al., 2007).  Using ArcGIS 9.3.2, we calculated the proportion of each habitat linkage within the major landcover classes, the proportion permanently conserved (GAP status 1-3; http://gapanalysis.nbii.gov/), the proportion within four habitat quality classes (i.e., avoided, marginal, sub-optimal, and preferred habitat) as defined in the landscape resistance model above, the road density within each habitat linkage, and the number of road crossings (defined as Class I-III roads completely crossing a linkage).  These landscape characteristics allowed for qualitative comparison between the modeled functional habitat linkage and the ad-hoc habitat linkage.

Results

Modeling Landscape Resistance

For each focal species, the modeled landscape resistance was low throughout the study area (Figure 3). Over 75% of the study area was predicted to be high quality habitat with low resistance and therefore highly traversable for all three species. 

Figure 3.  Resistance surfaces for three focal species (a) black bear, b) bobcat, and c) fisher) developed using literature review and expert opinion.  Light to dark red indicates high resistance to movement.  Yellow to green indicates lower resistance (high permeability) to movement.

Figure 3. Resistance surfaces for three focal species, a) black bear, b) bobcat, and c) fisher, developed using literature review and expert opinion. Light to dark red indicates high resistance to movement. Yellow to green indicates lower resistance (high permeability) to movement.

Identifying Habitat Linkages

Predicted habitat linkages for all three species, based on the species-specific landscape resistance surfaces, followed a route from the Jay and Giant Wilderness Block to Split Rock Wild Forest, and were merged to create a single functional habitat linkage (Figure 4).  The functional habitat linkage is 21 km in length, running northeast from Giant Mountain Wilderness Area to Split Rock Wild Forest, with an average width of 2 km.  The total area included in the functional habitat linkage is 93 km2.  It crosses Route I-87 southeast of Elizabethtown, NY and then passes through a large, roadless area before narrowing to 800 m outside of Westport, NY.

Figure 4.  Minimum cost-distance habitat linkages for a) black bear, b) bobcat, and c) fisher.  These were merged to create the functional habitat linkage d).

Figure 4. Minimum cost-distance habitat linkages for a) black bear, b) bobcat, and c) fisher. These were merged to create the functional habitat linkage d).

Alternate Scenarios

We produced ten black bear or twelve bobcat and fisher alternate scenario resistance surface models depending on the number of landscape factors included.  For all three species, the agreement between alternate scenarios and the original landscape resistance model was very high (Table 3).  Kappa agreement was substantial to almost perfect for most alternate scenarios; in only one scenario for fisher, compressed permeability score inputs for canopy cover resulted in moderate agreement.  However, when the landscape permeability score inputs were compressed for all landscape factors in concert for all three species, there was less agreement between landscape resistance surface maps (4-52% overlap).  These results suggest the models are insensitive to changes in assigned permeability scores of any one parameter but are sensitive to scenarios that diminish the differences between classes of all landscape factors.  In other words, results differed most from the original resistance model when parameter estimates underemphasized a focal species’ perception of the difference in resistance across a landscape (Beier et al., 2009).

Table 3.  Summary of the Kappa statistic pixel-by-pixel comparison of alternate landscape resistance surfaces for each focal species with the original landscape resistance surface model for each species showing the Kappa statistic value, lower confidence interval (LCI), upper confidence interval (UCI), and the agreement interpretation, as well as the percent overlap of habitat linkages predicted by the alternate scenarios with the habitat linkage predicted from the original model.

Table 3. Summary of the Kappa statistic pixel-by-pixel comparison of alternate landscape resistance surfaces for each focal species with the original landscape resistance surface model for each species showing the Kappa statistic value, lower confidence interval (LCI), upper confidence interval (UCI), and the agreement interpretation, as well as the percent overlap of habitat linkages predicted by the alternate scenarios with the habitat linkage predicted from the original model.

Species-specific habitat linkages generated by CorridorDesigner using the alternate scenario resistance surfaces generally overlapped with the functional habitat linkage.  The modeled habitat linkage location for black bear remained stable as factor weights and permeability scores were changed, with greater than 50% overlap in all scenarios (Table 3).  Habitat linkage locations for fisher tended to be stable; however, when all parameters were compressed, the habitat linkage location overlapped by only 7%.  Bobcat habitat linkage locations predicted by alternate resistance surfaces were more variable.  The lowest overlaps occurred when the permeability scores for the landscape factor representing core and edge habitats was changed or when all landscape factor scores were compressed, indicating that the location of bobcat habitat linkages is sensitive to the core/edge habitat parameter and to overestimates of the importance of each landscape factor class.

Figure 5. Visual representation of a) compressed and b) dispersed permeability score scenarios for black bear landscape factor “distance to roads.”  The bars indicate the plausible biological range for each class within the landscape factor and the boxes indicate the permeability score.  Note that the compressed permeability scores are more similar across all classes while the dispersed permeability scores are more spread out than the c) original permeability scores.

Figure 5. Visual representation of a) compressed and b) dispersed permeability score scenarios for black bear landscape factor “distance to roads.” The bars indicate the plausible biological range for each class within the landscape factor and the boxes indicate the permeability score. Note that the compressed permeability scores are more similar across all classes while the dispersed permeability scores are more spread out than the c) original permeability scores.

Comparison to Ad-hoc Habitat Linkage

Only 4% of the functional habitat linkage intersects with the ad-hoc habitat linkage (Figure 6).  In comparison to the previous ad-hoc habitat linkage, the functional habitat linkage has a lower road density, larger area, and shorter length (Table 4).  The amount of permanent conservation land (GAP status 1-3; http://gapanalysis.nbii.gov/) is greater in the functional habitat linkage, both in total acres conserved and percent of the total habitat linkage conserved.  The distribution of habitat quality for the three focal species in both habitat linkages tends toward preferred habitat (Figure 5).  In all cases, the functional habitat linkage has a higher proportion of preferred and highly traversable habitat.  Conversely, the ad-hoc habitat linkage has a higher proportion of avoided habitat for all three species. 

 

Figure 6.  The functional habitat linkage generated by minimum cost-distance and landscape resistance models is located south of the ad-hoc habitat linkage (Northeast Wilderness Trust, 2009).

Figure 6. The functional habitat linkage generated by minimum cost-distance and landscape resistance models is located south of the ad-hoc habitat linkage (Northeast Wilderness Trust, 2009).

 

 

Table 4.  Summary of landscape characteristics comparing the Split Rock Wildway Planning Area, functional habitat linkage, and ad-hoc habitat linkage.  High and low quality habitat percentages were calculated using the landscape resistance models from this study, conserved land percentage was calculated using the Protected Areas Database and the National Conservation Easement Database, and land cover type percentages were calculated from the NOAA C-CAP dataset (NCED, 2011; USGS National Gap Analysis Program, 2011).

Table 4. Summary of landscape characteristics comparing the Split Rock Wildway Planning Area, functional habitat linkage, and ad-hoc habitat linkage. High and low quality habitat percentages were calculated using the landscape resistance models from this study, conserved land percentage was calculated using the Protected Areas Database and the National Conservation Easement Database, and land cover type percentages were calculated from the NOAA C-CAP dataset (NCED, 2011; USGS National Gap Analysis Program, 2011).

 

Discussion

The functional habitat linkage generated by this study gives one option for prioritizing areas that could ensure long-term connectivity for wide-ranging mammals within the SRW between the Adirondack Mountains and the Champlain Valley.  For three focal species, the landscape resistance models indicated that there is a high-level of connectivity and high-quality habitat for the three focal species in the mostly forested landscape of the SRW.  Minimum cost-distance models predicted a functional habitat linkage across a southern route from Giant Wilderness Area to Split Rock Wild Forest suggesting the best location for a habitat linkage intended to maintain connectivity for black bear, bobcat, and fisher based on this dataset and weighting scheme (Figure 6).  Areas outside the SRW, and outside the scope of this paper, may also provide connectivity for these species between the Adirondack Mountains and the Green Mountains (Long, 2007; Zeh & Marangolo, 2010).  This study area was chosen to provide a comparison between a functional habitat linkage and an existing ad-hoc habitat linkage plan.

Of particular note to conservation planners is that the functional habitat linkage had very little overlap with an ad-hoc habitat linkage planned for the northern part of the study area.  Additionally, the functional habitat linkage has a larger area presently conserved and less area in land-use types (e.g. developed and agricultural land) that may require mitigation in order to remain permeable to wildlife movement.  The functional habitat linkage is larger which could imply that it would be more expensive to conserve.  However, the combination of more area presently conserved and a smaller area in less permeable land-use types suggests that the functional habitat linkage provides a more feasible option for maintaining connectivity across the SRW.  Further studies should focus on validation of the models presented in this study.

Validation of model results is often difficult and costly (Beier et al., 2008; Sawyer et al., 2011). Uncertainty analyses (cf. Beier et al. 2009, Ray & Burgman 2006) capture a range of model outcomes given parameters defined by data and/or expert knowledge.  Multiple model simulations using varied parameter estimates reflecting the uncertainty of our model tended to converge on the same functional habitat linkage.  Our results suggest that the predicted functional habitat linkage should be included in conservation plans aimed at protecting the existing connectivity between the High Peaks of the Adirondacks and Lake Champlain.  Maintaining landscape connections between the Split Rock Wild Forest and the larger wild areas of the Adirondack Park may provide a stepping stone connecting the Adirondack Park with the greater Northern Appalachian/Acadian ecoregion thereby adding to a network of core reserves, matrix lands, and corridors believed to be critical to the long-term health of both the Adirondack Park and the greater ecoregion (Trombulak et al., 2008).

We have presented one example of systematic conservation planning that incorporates scientific understanding of focal species and their needs in a particular landscape and that is a defensible alternative to ad-hoc conservation planning.  However, it is important to recognize the assumptions and specific objectives of the analysis.  Our analysis models the best habitat linkage for the three focal species based on the available data.  While the vagility and large area requirements of these species make it possible for them to serve as umbrella species (i.e., protecting habitat for them would have the side benefit of protecting habitat for many other facets of biodiversity), the inclusion of other focal species or focal ecosystem characteristics could result in a different habitat linkage (Roberge & Angelstam, 2004; Thorne et al., 2006).  For example, the ad-hoc linkage has a larger wetland area and might overlap with a functional habitat linkage based on the requirements of wetland obligate species.  The functional habitat linkage presented here can provide a guiding tool for organizations determining where best to spend their financial and human resources if connectivity of wide-ranging mammals is a priority.  Organizations that prioritize other conservation values can use similar analyses to generate systematic conservation plans incorporating those goals, including recreational access, aesthetic values, and water quality (Margules & Pressey, 2000).

Conclusions

Connected landscapes provide wildlife and natural processes the flexibility to respond to both natural and human-created disturbances.  While most researchers agree that the identification of habitat linkages should be based on empirical evidence such as movement of radio-tagged animals, population studies, or estimates of genetic distance, the availability of empirical data is limited and often expensive (Bates & Jones, 2007; Brown et al., 2010; Crooks & Sanjayan, 2006a; Feinberg, 2007; Long, 2007).  The modeling framework used in this study provides an example of how land managers can incorporate existing knowledge and GIS data to create science-based functional connectivity models at a scale useful to local conservation planning.  It allows and encourages conservation planners to be explicit about what they are trying to achieve and intentional about the information that they use to develop conservation plans.  The modeling framework also provides insight into conservation values that highlight areas not previously considered high priority and allows for transparent consideration of multiple alternatives.  With the advent of easy-to-use and low cost toolkits, these analyses help us to promote accountability in decision-making, transparency in conservation decisions, and efficiency in the allocation of financial and human resources.

Acknowledgments 

The authors thank the Northeast Wilderness Trust for providing funding for this study. We would also like to thank Dr. Therese M. Donovan and Dr. Cathy Paris for their comments on earlier drafts of this work as well as the reviewers from AJES for providing comments that substantially improved this work.

Authors

Rose A. Graves1,2*, Deane Wang1

*Corresponding author email: ragraves@wisc.edu

1 Ecological Planning Program, University of Vermont, Burlington, VT

2Current contact: Ecosystem and Landscape Ecology Lab, University of Wisconsin-Madison

 Rose A. Graves is a Ph.D. student at the University of Wisconsin-Madison and the former Conservation Coordinator at Northeast Wilderness Trust.  Deane Wang, Ph.D. is Associate Professor of Natural Resources in the Ecological Planning Program at the University of Vermont.

References

 Abouelezz, H. G. 2009. Quantifying bobcat (Lynx rufus) movement patterns and habitat selection in Northwestern Vermont. University of Vermont, Burlington, VT. 

Adriaensen, F., J. P. Chardon, G. De Blust, E. Swinnen, S. Villalba, H. Gulinck, and E. Matthysen. 2003. The application of ‘least-cost’ modelling as a functional landscape model. Landscape and Urban Planning 64:233-247. 

Allen, A. W. 1983. Habitat suitability index models: Fisher. Page 19. U.S. Fish and Wildlife Service. 

Arthur, S. M. 1987. Ecology of fishers in southcentral Maine. University of Maine, Orono, ME. 

Arthur, S. M. and W. B. Krohn. 1991. Activity patterns, movements, and reproductive ecology of fishers in southcentral Maine. Journal of Mammalogy 72:379-385. 

Aubry, K. B. and D. B. Houston. 1992. Distribution and status of the fisher in Washington. Northwestern Naturalist 73:69-79. 

Austin, J. M., K. Viani, F. Hammond, and C. Slesar. 2005. A GIS-based identification of potentially significant wildlife habitats associated with roads in Vermont. 

Baldwin, R. F., R. M. Perkl, S. C. Trombulak, and W. B. Burwell. 2010. Modeling Ecoregional Connectivity. Pages 349-367 in S. C. Trombulak and R. F. Baldwin, editors. Landscape-scale Conservation Planning. Springer Netherlands. 

Barnum, S., K. Rinehart, and M. Elbroch. 2007. Habitat, Highway Features, and Animal-Vehicle Collision Locations ad Indicators or Wildlife Crossing Hotspots. Pages 511-518 in C. L. Irwin, D. Nelson, and K. P. McDermott, editors. International Conference on Ecology and Transportation. Center for Transporation and the Environment, North Carolina State University, Raleigh, NC. 

Bates, W. and A. Jones. 2007. Least-cost corridor analysis for evaluation of lynx habitat connectivity in the middle Rockies. Utah Office of the Nature Conservancy. 

Bateson, E. M. 2005. Two Countries, One Forest: Launching a Landscape-Scale Conservation Collaborative in the Northern Appalachian Region of the United States and Canada. George Wright Forum 22:35-45. 

Beier, P., D. Majka, and W. D. Spencer. 2008. Forks in the road: choices in procedures for designing wildland linkages. Conservation Biology 22:834-851. 

Beier, P., D. R. Majka, and S. L. Newell. 2009. Uncertainty analysis of least-cost modeling for designing wildlife linkages. Ecological Applications 19:2067-2077. 

Beier, P. and R. F. Noss. 1998. Do Habitat Corridors Provide Connectivity? Conservation Biology 12:1241-1252. 

Beier, P., K. Penrod, C. Luke, W. Spencer, and C. Cabanero. 2006. South Coast Missing Linkages: restoring connectivity to wildlands in the largest metropolitan area in the United States. Pages 555-586 in K. R. Crooks and M. Sanjayan, editors. Connectivity Conservation. 

Bennett, A. F. 2003. Linkages in the Landscape: The Role of Corridors and Connectivity in Wildlife Conservation. IUCN, Gland, Switzerland and Cambridge, UK. 

Beringer, J. J., S. G. Siebert, and H. R. Pelton. 1990. Incidence of road crossing by black bears in Pisgah National Forest, North Carolina. International Conference on Bear Research and Managment. 8:85-92. 

Berke, P. R. 2007. Ecology and new direction for land use planning: barriers and opportunities to change. Pages 59-70 in R. Kihslinger and J. Wilkinson, editors. Lasting landscapes: reflections on the role of conservation science in land use planning. Environmental Law Institute, Washington, D.C. 

Boileau, F., M. Crete, and J. Huot. 1994. Food Habits of the Black Bear, Ursus americanus, and habitat use in Gaspesie Park, Eastern Quebec. Canadian Field-Naturalist 108:162-169. 

Boyle, K. A. and T. T. Fendley. 1987. Habitat Suitability Index Models: Bobcat. Page 16. U.S. Department of Interior Fish and Wildlife Service. 

Brandenburg, D. M. 1996. Effects of roads on behavior and survival of black bears in coastal North Carolina. University of Tennessee, Knoxville, TN. 

Brody, A. J. and M. R. Pelton. 1989. Effects of roads on black bear movement in western North Carolina. Wildlife Society Bulletin 17:5-10. 

Brown, M., C. Cheeseman, L. Garrett, T. Dunham, and D. Bryant. 2010. Adirondack – Tug Hill Connectivity Project Planning Phase – Final Report. The Nature Conservancy. 

Buck, S. G., C. Mullis, A. S. Mossman, I. Show, and C. Coolahan. 1994. Habitat use by fishers in adjoining heavily and lightly harvested forest. Pages 283-296 in S. W. Buskirk, A. S. Harestad, M. G. Raphael, and R. A. Powell, editors. Martens, sables, and fishers: biology and conservation. Cornell University Press, Ithaca, NY. 

Buskirk, S. W. and R. A. Powell. 1994. Habitat ecology of fishers and American marten. Pages 283-296 in S. W. Buskirk, A. S. Harestad, M. G. Raphael, and R. A. Powell, editors. Martens, sables, and fisher: biology and conservation. Cornell University Press, Ithaca, NY. 

Calabrese, J. M. and W. F. Fagan. 2004. A Comparison-Shopper’s Guide to Connectivity Metrics. Frontiers in Ecology and the Environment 2:529-536. 

Carroll, C. 2010. Connectivity Analysis Toolkit (CAT) Manual, Version 1.1. Klamath Center for Conservation Research. 

Chetkiewicz, C.-L. B., C. C. St. Clair, and M. S. Boyce. 2006. Corridors for Conservation: Integrating Pattern and Process. Annual Review of Ecology, Evolution, and Systematics 37:317-342. 

Clark, J. D., J. E. Dunn, and K. G. Smith. 1993. A Multivariate Model of Female Black Bear Habitat Use for a Geographic Information System. Journal of Wildlife Management 57:519-526. 

Clevenger, A. P., J. Wierzchowski, B. Chruszcz, and K. Gunson. 2002. GIS-generated, expert-based models for identifying wildlife habitat linkages and planning mitigation passages. Conservation Biology 16:503-514. 

Costello, C. M. 1992. Black bear habitat ecology in the central Adirondacks as related to food abundance and forest management. State University of New York, College of Environmental Science and Forestry, Syracuse, NY. 

Coulter, M. W. 1966. Ecology and management of fishers in Maine. SUNY ESF, Syracuse, NY. 

Crooks, K. R. 2002. Relative Sensitivities of Mammalian Carnivores to Habitat Fragmentation. Conservation Biology 16:488-502. 

Crooks, K. R. and M. Sanjayan, editors. 2006a. Connectivity Conservation. Cambridge University Press, Boston, MA. 

Crooks, K. R. and M. Sanjayan. 2006b. Connectivity conservation: maintaining connections for nature.in K. R. Crooks and M. Sanjayan, editors. Connectivity Conservation. Cambridge University Press.

Dark, S. J. 1997. A landscape-scale analysis of mammalian carnivore distribution and habitat use by fisher. Humboldt State University, Arcata, CA. 

deVos, A. 1952. Ecology and management of fisher and marten in Ontario. Page 90. Ontario Department of Lands and Forests. 

Dixon, J. D., M. K. Oli, M. C. Wooten, T. H. Eason, J. W. McCown, and D. Paetkau. 2006. Effectiveness of a regional corridor in connecting two Florida black bear populations. Conservation Biology 20:155-162. 

Driezen, K., F. Adriaensen, C. Rondinini, C. P. Doncaster, and E. Matthysen. 2007. Evaluating least-cost model predictions with empirical dispersal data: A case-study using radiotracking data of hedgehogs (Erinaceus europaeus). Ecological Modelling 209:314-322. 

Elowe, K. D. and W. E. Dodge. 1989. Factors affecting black bear reproductive success and cub survival. J. Wildlife Management 53:962-968. 

Ewers, R. M. and R. K. Didham. 2006. Confounding factors in the detection of species responses to habitat fragmentation. Biological Reviews 81:117-142. 

Fecske, D. M., R. E. Barry, F. L. Precht, H. B. Quigley, S. L. Bittner, and T. Webster. 2002. Habitat use by female black bears in western Maryland. Southeastern Naturalist 1:77-92. 

Feinberg, J. 2007. Conserving the Connections: A Nationwide Inventory of State-based Habitat Connectivity Analysis. Road Ecology Center, John Muir Institute of the Environment, UC Davis. 

Fisher, J. T. 2004. Populating the wildlife habitat suitability component of the ALCES Model: Fisher, moose, and black bear in Northeastern Alberta. Page 22. Alberta Research Council. 

Fox, L. B. 1990. Ecology and population biology of the bobcat, Felis rufus in New York. State University of New York, Syracuse. 

Giles, R. H., Jr. 1998. Natural Resource Management Tomorrow: Four Currents. Wildlife Society Bulletin 26:51-55. 

Groves, C. R., D. B. Jensen, L. L. Valutis, K. H. Redford, M. L. Shaffer, J. M. Scott, J. V. Baumgartner, J. V. Higgins, M. W. Beck, and M. G. Anderson. 2002. Planning for Biodiversity Conservation: Putting Conservation Science into Practice. BioScience 52:499-512. 

Haddad, N. M., D. K. Rosenberg, and B. R. Noon. 2000. On Experimentation and the Study of Corridors: Response to Beier and Noss. Conservation Biology 14:1543-1545. 

Hall, E. R. 1981. The Mammals of North America. John Wiley and Sons, New York, NY. 

Hammond, F. M. 2002. The effects of resort and residential development on black bears in Vermont. Page 113. Vermont Agency of Natural Resources, Dept of Fish and Wildlife. 

Hansen, K. 2007. Bobcat: master of survival. Oxford University Press, Oxford; New York. 

Hanski, I. 1999. Habitat Connectivity, Habitat Continuity, and Metapopulations in Dynamic Landscapes. Oikos 87:209-219. 

Hargrove, W. W., F. M. Hoffman, and R. A. Efroymson. 2004. A practical map-analysis tool for detecting potential dispersal corridors. Landscape Ecology 20:361-373. 

Harrison, R. L. 1992. Toward a Theory of Inter-Refuge Corridor Design. Conservation Biology 6:293-295. 

Heinemeyer, K. S. 1993. Temporal dynamics in the movements, habitat use, activity, and spacing of reintroduced fishers in northwestern Montana. University of Montana, Missoula, MT. 

Hilty, J. A., W. Z. Lidicker Jr., and A. M. Merenlender, editors. 2006. Corridor Ecology: the Science and Practice of Linking Landscapes for Biodiversity Conservation. Island Press, Washington D.C. 

Hilty, J. A. and A. M. Merenlender. 2004. Use of Riparian Corridors and Vineyards by Mammalian Predators in Northern California. Conservation Biology 18:126-135. 

Huck, M., W. Jedrzejewski, T. Borowik, B. Jedrzejewska, S. Nowak, and R. Myslajek. 2011. Analyses of least cost paths for determining effects of habitat types on landscape permeability: wolves in Poland. Acta Theriologica 56:91-101. 

Hugie, R. D. 1982. Black bear ecology and management in the northern conifer-deciduous forests of Maine. University of Montana, Missoula, MT. 

Johnson, C. J. and M. P. Gillingham. 2004. Mapping Uncertainty: Sensitivity of Wildlife Habitat Ratings to Expert Opinion. Journal of Applied Ecology 41:1032-1041. 

Johnson, S. A., H. D. Walker, and C. M. Hudson. 2010. Dispersal Characteristics of Juvenile Bobcats in South-Central Indiana. Journal of Wildlife Management 74:379-385. 

Jones, J. L. and E. O. Garton. 1994. Selection of successional stages by fishers in northcentral Idaho. Pages 377-387 in S. W. Buskirk, A. S. Harestad, and M. G. Raphael, editors. Martens, sables, and fishers: biology and conservation. Cornell University Press, Ithaca, NY. 

Kart, J., R. Regan, S. R. Darling, C. Alexander, K. Cox, M. Ferguson, S. Parren, K. Royar, and B. Popp, editors. 2005. Vermont’s Wildlife Action Plan. Vermont Fish and Wildlife Department, Waterbury, VT. 

Kasworm, W. F. and T. L. Manley. 1990. Road and trail influences on grizzly bears and black bears in northwest Montana. International Conference on Bear Research and Managment 8. 

Kelly, G. M. 1977. Fisher (Martes pennanti) biology in the White Mountain National Forest and adjacent areas. University of Massachusetts, Amherst, MA. 

Kindlmann, P. and F. Burel. 2008. Connectivity measures: a review. Landscape Ecology 23:879-890. 

King, J. E. 2004. Software solutions for obtaining a kappa-type statistic for use with multiple raters. Southwest Educational Research Association, Dallas, TX. 

Kirk, T. A. and W. J. Zielinski. 2010. Functional habitat connectivity of the American marten in northeastern California using least-cost corridor modeling. USDA Forest Service. 

Kolowski, J. M. and A. Woolf. 2002. Microhabitat Use by Bobcats in Southern Illinois. The Journal of Wildlife Management 66:822-832. 

Krohn, W. B. 2010. Personal communication. Maine Cooperative Fish and Wildlife Group, Orono, ME. 

Lancaster, P. A., J. Bowman, and B. A. Pond. 2008. Fisher, farms, and forests. Environmental Management 42:93-101. 

Landers, J. L., R. J. Hamilton, A. S. Johnson, and R. L. Marchinton. 1979. Foods and habitat of black bears in southeastern North Carolina. J. Wildlife Management 43:143-153. 

LaPoint, S. D. 2007. Mammalian road crossing patterns along a major highway within Adirondack Park of New Yor State, USA. State University of New York, College of Environmental Science and Forestry, Syracuse, NY. 

LaPoint, S. D. and R. W. Kays. 2011. Behavior and ecology of fisher (Martes pennanti) inhabiting an urban ecosystem. American Society of Mammalogists, Portland, OR. 

Lariviere, S. 2001. Ursus americanus. Mammalian Species 647:1 – 11. 

Lentz, W. M., R. L. Marchinton, and D. M. Carlock. 1980. Black bear habitat in north Georgia: some implications of wilderness designation. Proc. Ann. Conf. S.E. Assoc. Fish and Wildl. Agencies 34:550-556. 

Levine, R., K. Yorita, M. Walsh, and M. Reynolds. 2009. A method for statistically comparing spatial distribution maps. International Journal of Health Geographics 8:7. 

Lindenmayer, D., R. J. Hobbs, R. Montague-Drake, J. Alexandra, A. Bennett, M. Burgman, P. Cale, A. Calhoun, V. Cramer, P. Cullen, D. Driscoll, L. Fahrig, J. Fischer, J. Franklin, Y. Haila, M. Hunter, P. Gibbons, S. Lake, G. Luck, C. MacGregor, S. McIntyre, R. Mac Nally, A. Manning, J. Miller, H. Mooney, R. Noss, H. Possingham, D. Saunders, F. Schmiegelow, M. Scott, D. Simberloff, T. Sisk, G. Tabor, B. Walker, J. Wiens, J. Woinarski, and E. Zavaleta. 2008. A checklist for ecological management of landscapes for conservation. Ecology Letters 11:78-91. 

Livaitis, J. A., J. T. Major, and J. A. Sherburne. 1987. Influence of season and human-induced mortality on spatial organization of bobcats (Felis rufus) in Maine. Journal of Mammalogy 68:100-106. 

Livaitis, J. A., J. A. Sherburne, and J. A. Bissonette. 1986. Bobcat habitat use and home range size in realtion to prey density. Journal of Wildlife Management 50:110-117. 

Long, R. A. 2006. Developing predicitive occurrence models for carnivores in Vermont using data collected with multiple noninvasive methods. University of Vermont, Burlington, VT. 

Long, R. A. 2007. Predicting occurrence and assessing connectivity for wide-ranging carnivores in the Greater Southern Lake Champlain Region of New York. The Nature Conservancy. 

Lovallo, M. J. and E. M. Anderson. 1996a. Bobcat (Lynx rufus) home range size and habitat use in northwest Wisconsin. American Midland Naturalist 135:241-252. 

Lovallo, M. J. and E. M. Anderson. 1996b. Bobcat movements and home ranges relative to roads in Wisconsin. Wildlife Society Bulletin 24:71-76. 

Maclachlan, W. S. 1981. Bobcat studies in New Hampshire. University of New Hampshire, Durham, N.H. 

Majka, D., J. Jenness, and P. Beier. 2007. CorridorDesigner: ArcGIS tools for designing and evaluating corridors. Available at http://corridordesign.org. 

Marangelo, P. 2010. Personal communication. Conservation ecologist, The Nature Conservancy, VT. 

Margules, C. R. and R. L. Pressey. 2000. Systematic conservation planning. Nature 405:243-253. 

McLaughlin, C. R. 1998. Modeling effects of food and harvest on female black bear populations. University of Maine, Orono, ME. 

McLaughlin, C. R., G. J. Matula Jr., and R. J. O’Connor. 1994. Synchronous reproduction by Maine black bears. International Conference on Bear Research and Managment. 1:471-479. 

Mitchell, M. S., J. W. Zimmerman, and R. A. Powell. 2002. Test of a habitat suitability index for black bears in the southern Appalachians. Wildlife Society Bulletin 30:794-808. 

Monserud, R. A. and R. Leemans. 1992. Comparing global vegetation maps with the Kappa statistic. Ecological Modelling 62:275-293. 

Morris, K. I. 1986. Bobcat assessment: Maine Department of Inland Fisheries and Wildlife. Page 22. State of Maine, Augusta. 

Morse, S. 2010. Personal communication. Keeping Track, Jericho, VT. 

NCED. 2011. National Conservation Easement Database. The Conservation Registry. http://www.conservationeasement.us/. 

New York State Department of Environmental Conservation. 2005. Split Rock Mountain Wild Forest Westport Boat Launch and Whallonsburg Fishing Access Site Unit Management Plan. NY DEC, Albany, NY. 

Northeast Wilderness Trust. 2009. Split Rock Wildway Conservaton Plan. Unpublished data, Bristol, VT. 

Noss, R. F. 1983. A regional landscape approach to maintain diversity. BioScience 33:700-706. 

Noss, R. F. 1987. Protecting natural areas in fragmented landscapes. Natural Areas Journal 7:2. 

Noss, R. F. 2003. A Checklist for Wildlands Network Designs. Conservation Biology 17:1270. 

NY DEC. 2007. Black Bears in New York: Natural history, range, and interactions with people.in W. Division of Fish, and Marine Resources, editor. New York State Department of Environmental Conservation, Albany, NY. 

Pinto, N. and T. Keitt. 2009. Beyond the least-cost path: evaluating corridor redundancy using a graph-theoretic approach. Landscape Ecology 24:253-266. 

Potter, D. N. 2002. Modelling fisher (Martes pennanti) habitat associations in Nova Scotia. Acadia University. 

Powell, R. A. 1993. The fisher: Life history, ecology, and behavior. University of Minnesota Press, Minneapolis, MN. 

Powell, R. A., S. W. Buskirk, and W. J. Zielinski. 2003. Fisher and Marten.in G. A. Feldhamer, B. C. Thompson, and J. A. Chapman, editors. Wild mammals of North America : biology, management, and conservation. Johns Hopkins University Press, Baltimore, MD. 

Powell, R. A., J. W. Zimmerman, D. E. Seaman, and J. F. Gilliam. 1996. Demographic analyses of a hunted black bear population with access to a refuge. Conservation Biology 10:224-234. 

Prato, T. and D. Fagre, editors. 2007. Sustaining Rocky Mountain Landscapes: Science, Policy, and Management for the Crown of the Continent Ecosystem. Resources for the Future, Washington, D.C. 

Ray, N. and M. A. Burgman. 2006. Subjective uncertainties in habitat suitability maps. Ecological Modelling 195:172-186. 

Reed, E. 2010. Personal Communication. Wildlife Biologist, New York State Department of Environmental Conservation. 

Riley, S. P. D., R. M. Sauvajot, T. K. Fuller, E. C. York, D. A. Kamradt, C. Bromley, and R. K. Wayne. 2003. Effects of Urbanization and Habitat Fragmentation on Bobcats and Coyotes in Southern California. Conservation Biology 17:566-576. 

Roberge, J.-M. and P. E. R. Angelstam. 2004. Usefulness of the Umbrella Species Concept as a Conservation Tool. Conservation Biology 18:76-85. 

Rogers, L. L. and A. W. Allen. 1987. Habitat suitability index models: Black bear, Upper Great Lakes Region. Page 54. U.S. Fish and Wildlife Service. 

Sawyer, S. C., C. W. Epps, and J. S. Brashares. 2011. Placing linkages among fragmented habitats: do least-cost models reflect how animals use landscapes? Journal of Applied Ecology. 

Schoen, J. 1990. Bear habitat management: A review and future perspective. Pages 143-154  International Conference on Bear Research and Management. 

Schooley, R. L., C. R. McLaughlin, W. B. Krohn, and G. J. Matula Jr. 1994. Spatiotemporal patterns of macrohabitat use by female black bears during fall. International Conference on Bear Research and Managment. 9:339-348. 

Shah, V. B. and B. H. McRae. 2008. Circuitscape: a tool for landscape ecology. Pages 62-66 in 7th Python in Science Conference. 

Simek, S. L. 1995. Impacts of forest land management on black bear populations in the central Adirondacks. State University of New York, College of Environmental Science and Forestry, Syracuse, NY. 

Singleton, P. H., W. L. Gaines, and J. F. Lehmkuhl. 2002. Landscape Permeability for Large Carnivores in Washington: A Geographic Information System Weighted-Distance and Least-Cost Corridor Assessment.in F. S. U.S. Department of Agriculture, editor. Pacific Northwest Research Station. 

Singleton, P. H. and J. F. Lehmkuhl. 2001. Using weighted distance and least-cost corridor analysis to evaluate regional-scale large carnivore habitat connectivity in Washington. Pages 583-594 in 2001 International Conference on Ecology and Transportation. Center for Transportation and the Environment, North Carolina State University, Raleigh, NC. 

Soule, M. and R. Noss. 1998. Rewilding and biodiversity: complementary goals for continental conservation. Wild Earth 8:18-28. 

Sunquist, M. E. and F. Sunquist. 2002. Wild Cats of the World. University of Chicago Press, Chicago, IL. 

Theobald, D. M., J. B. Norman, and M. R. Sherburne. 2006. FunConn v1 User’s Manual: ArcGIS tools for Functional Connectivity Modeling. Natural Resource Ecology Lab, Colorado State University, Fort Collins, CO. 

Thorne, J. H., D. Cameron, and J. F. Quinn. 2006. A Conservation Design for the Central Coast of California and the Evaluation of Mountain Lion as an Umbrella Species. Natural Areas Journal 26:137-148. 

Tigas, L. A., D. H. Van Vuren, and R. M. Sauvajot. 2002. Behavioral responses of bobcats and coyotes to habitat fragmentation and corridors in an urban environment. Biological Conservation 108:299-306. 

Trombulak, S. C., M. G. Anderson, R. A. Baldwin, K. Beazley, J. C. Ray, C. Reining, G. Woolmer, C. Bettigole, G. Forbes, and L. Gratton. 2008. The Northern Appalachian/Acadian Ecoregion: Priority Locations for Conservation Action. Two Countries, One Forest. 

Tucker, S. A., W. R. Clark, and T. E. Gosselink. 2008. Space use and habitat selection by bobcats in the fragmented landscape of south-central Iowa. Journal of Wildlife Management 72:1114-1124. 

Urban, D. L. and T. H. Keitt. 2001. Landscape connectivity: a graph-theoretic perspective. . Ecology 82:1205-1218. 

USGS National Gap Analysis Program. 2011. Protected Areas Database of the United States (PADUS) version 1.2. Download or view data from: USGS Gap Analysis Program website (www.gapanalysis.usgs.gov/PADUS). 

Viera, A. J. and J. M. Garrett. 2005. Understanding interobserver agreement: the kappa statistic. Fam Med 37:360-363. 

Villarrubia, C. R. 1982. Movement ecology and habitat utilization of black bears in Cherokee National Forest, Tennessee. University of Tennessee, Knoxville, TN. 

Walker, R. and L. Craighead. 1997. Analyzing movement corridors in Montana using GIS.in 1997 International ESRI Users conference. Environmental Sciences Research Institute. 

Walker, R. S., A. J. Novaro, and L. C. Branch. 2007. Functional connectivity defined through cost-distance and genetic analyses: a case study for the rock-dwelling mountain vizcacha (Lagidium viscacia) in Patagonia, Argentina. Landscape Ecology 22:1303-1314. 

Woolf, A., C. K. Nielsen, T. Weber, and T. J. Gibbs-Kieninger. 2002. Statewide modeling of bobcat, Lynx rufus, habitat in Illinois, USA. Biological Conservation 104:191-198. 

Zeh, D. and P. Marangolo. 2010. FunConn Analysis of Habitat Connectivity between the Green Mountains, VT and the Adirondack Mountains, NY. The Nature Conservancy.

 Appendix

Table 2a. Summary of landscape factor classes and the black bear permeability scores assigned to each class (i.e., 0 indicates a completely resistant or completely unusable class and 100 represents completely traversable, optimal habitat) including the biologically plausible range for each score as well as the compressed and dispersed values used to create alternate scenarios.

Appendix A1. Summary of landscape factor classes and the black bear permeability scores assigned to each class (i.e., 0 indicates a completely resistant or completely unusable class and 100 represents completely traversable, optimal habitat) including the biologically plausible range for each score as well as the compressed and dispersed values used to create alternate scenarios.

 

Table 2b.  Summary of landscape factor classes and the bobcat permeability scores assigned to each class (i.e., 0 indicates a completely resistant or completely unusable class and 100 represents completely traversable, optimal habitat) including the biologically plausible range for each score as well as the compressed and dispersed values used to create alternate scenarios.

Appendix A2. Summary of landscape factor classes and the bobcat permeability scores assigned to each class (i.e., 0 indicates a completely resistant or completely unusable class and 100 represents completely traversable, optimal habitat) including the biologically plausible range for each score as well as the compressed and dispersed values used to create alternate scenarios.

 

Appendix A3.  Summary of landscape factor classes and the fisher permeability scores assigned to each class (i.e., 0 indicates a completely resistant or completely unusable class and 100 represents completely traversable, optimal habitat) including the biologically plausible range for each score as well as the compressed and dispersed values used to create alternate scenarios.

Appendix A3. Summary of landscape factor classes and the fisher permeability scores assigned to each class (i.e., 0 indicates a completely resistant or completely unusable class and 100 represents completely traversable, optimal habitat) including the biologically plausible range for each score as well as the compressed and dispersed values used to create alternate scenarios.

 

No comments to display

Leave a Comment



You must be logged in to post a comment.