Search

Landscape connectivity is a key component for successful dispersal of wildlife. Many approaches to quantify landscape connectivity utilize landscape characteristics and wildlife behavior in response to those characteristics, whereas other approaches focus on wildlife dispersal behavior. Combining landscape structure and wildlife behavior (e.g. movement, food acquisition, response to predation risk) provides critical information for conservation and management of many species. Individual-based models (IBM) are useful tools for evaluating the interaction between landscape characteristics and individual behaviors, thereby providing insights into management and conservation. Our objective was to evaluate the effect of reductions in landscape connectivity on northern flying squirrel dispersal movements under alternative timber management scenarios within the modeling framework of a spatially-explicit IBM. We simulated timber harvests of 6% and 9% of old-growth forest, the primary habitat for this species, distributed across 5, 10 or 15 harvest locations. We measured the influence of these scenarios upon landscape connectivity for flying squirrels. Landscapes with greater connectivity exhibited longer flying squirrel dispersal distances, more sinuous dispersal paths, and a greater total area of landscape utilization. However, connectivity was not directly correlated with habitat loss. The harvest scenario with 6% harvest of old growth distributed among 15 harvest locations had the lowest connectivity index value despite other scenarios modeling a greater percentage of old growth loss. The IBM demonstrated the importance of behaviors such as path tortuosity and movement rates in conjunction with landscape configuration in influencing the movement of dispersing individuals. The spatially explicit IBM provided a framework to evaluate connectivity from a fine-scale behavioral rather than structural perspective as well as to evaluate the distribution of new home range locations, which could be a useful management tool when evaluating the influence of landscape heterogeneity and stochastic behavior on wildlife movement and dispersal.

We evaluated the northern flying squirrel (Glaucomys sabrinus griseifrons), a known associate of key habitat features and processes of old-growth forest, for its capacity as a broad-scale ecological indicator of temperate rainforest ecosystem condition in southeastern Alaska, USA. We utilized a spatially explicit, resource-selection function to evaluate its distribution relative to landscape composition and structure at local (within home-range) and broad (home-range selection) spatial scales, followed by a moving-window analysis to model patch occupancy across this landscape. We found strong support for the influence of type, size, and compositional elements: large, old-growth patches were selected at both spatial scales, and regenerating forest patches ≤40 yrs old were selected against at the broader scale. More importantly, we found that occupancy was related to critical thresholds in composition: patches required ≥73% old-growth forest cover or a minimum total area of 73 ha of old-growth forest to be occupied by flying squirrels. A non-uniform pattern of selection for patches with higher structural connectivity at the local scale, but not at the broader scale, was likely a result of the “empty forest” phenomenon, in which remnant patches were inaccessible in this fragmented landscape. These results are consistent with recent studies of this and related species and suggest that occurrence of northern flying squirrels in southeastern Alaska is influenced by a number of landscape structure and compositional variables that relate critically to late-seral forest conditions. These findings shed new light on the utility of this species as an ecological indicator and its functional relevance to the resilience of fragmented forest ecosystems.

In many regions of the world, forest management has shifted from practices emphasizing timber production to more sustainable harvesting that integrates ecological values, including maintenance of biodiversity, wildlife habitat, and ecological goods and services. To this end, management strategies emphasize retention of stand structures that meet the needs of forest-obligate wildlife species and enhance connectivity across landscapes. However, little is known about the effects on arboreal rodents of varying the amount or spatial distribution of retained structures. We quantified the responses of northern flying squirrels ( Glaucomys sabrinus ) to retention harvests of varying levels (15%, 40%, 75%, and 100% of original basal area) and spatial patterns (trees uniformly dispersed vs. aggregated in 1-ha patches), using six experimental treatments replicated at three locations in southwestern Oregon and Washington. Relative abundance of northern flying squirrels decreased following harvest; minimum number of squirrels known alive (MNKA) in the control (100%) and 75% retention treatment was significantly higher than in the 15% or 40% treatments. In mixed-effects regression models, MNKA increased with treatment-unit basal area and amount of surrounding mature (>80-year-old) forest, suggesting that squirrel abundance was influenced by local structure and landscape-scale variables. However, only basal area contributed to best-fit models of reproductive female abundance. Our results suggest a threshold response of northern flying squirrels to green-tree retention somewhere between 40% and 75% that is likely to be influenced by the spatial pattern of retention and landscape context. This study underscores previous conclusions that northern flying squirrels are sensitive to logging at both local and larger landscape scales, and demonstrates the current minimum retention standard of 15% will not provide suitable habitat for this species.

Habitat reserves are a common strategy used to ensure viability of wildlife populations and communities. The efficacy of reserves, however, is rarely empirically evaluated. We examined the likelihood that small (650 ha), isolated habitat reserves composed of old-growth Sitka spruce ( Picea sitchensis )–western hemlock ( Tsuga heterophylla ) rain forest (upland-OG) and mixed-conifer peatlands (peatland-MC) would sustain populations of northern flying squirrels ( Glaucomys sabrinus ) in the absence of immigration or emigration within the Tongass National Forest in Southeast Alaska. We used demographic data obtained from a study of flying squirrels on Prince of Wales Island in Southeast Alaska and litter size from flying squirrels in similar habitat to estimate per capita rate of increase ( r ) of flying squirrels in upland-OG ( r = 0.14, SD = 0.42) and peatland-MC habitats ( r = 0.01, SD = 0.39). Our results indicated that peatland-MC habitat was unlikely to sustain populations and that viability of flying squirrel populations in small habitat reserves largely depended on the upland-OG forest component. We subsequently estimated time to extinction ( T N ) based on r , its variance ( v ), and the potential population ceiling ( K ). We used T N to calculate the probabilities ( P t ) that squirrel populations would persist in small reserves containing 100%, 50%, and 25% upland-OG habitat for 25, 50, and 100 years. In each scenario, we calculated T N and P t for 2 levels of v . For the best-case scenario (100% upland-OG forest, lowest variance, t = 25 years), T N was 507 years and P t was 0.95. For the worst-case scenario (25% upland-OG forest, highest variance, t = 100 years), T N was 237 years and P t was 0.66. Minimum patch size of upland-OG forest required for a high probability ( P t = 0.95) of sustaining a flying squirrel population in isolation with relatively low demographic variability ( v = 0.34) for 25, 50, or 100 years was 578, 5077, and 78,935 ha, respectively. We concluded that it was unlikely that small isolated habitat reserves could sustain populations of flying squirrels for >25 years in the absence of immigration. Consequently, dispersal among small reserves is critical to ensure that they function to support metapopulations of northern flying squirrels.

Whether land management planning provides for sufficient habitat to sustain viable populations of indigenous wildlife is one of the greatest challenges confronting resource managers. Analyses of the effects of land management on natural resources often rely on qualitative assessments that focus on single species to reflect the risk of wildlife extinction across a planning area. We propose a conceptual framework for sustainable management of wildlife habitat that explicitly acknowledges the greater risk of an extinction event when considering the viability of multiple species, e.g., an indigenous vertebrate fauna. This concept is based on the principle that the likelihood of at least one event (i.e., species extinction) is the joint probability of the extinction probabilities of individual species, assuming independence among species` responses to disturbance. We present an ecological rationale to support the view that, at a spatial scale of 104-106 ha (i.e., planning area) and a temporal scale of 102 years (i.e., planning horizon), wildlife species operating at varying ecological scales respond relatively independently to disturbances typically associated with land management. We use a hypothetical scenario of a wildlife viability assessment and Monte Carlo simulation to demonstrate that the probability of `any extinction` is consistently higher than the probability of the `single most likely` extinction, and that the difference between these values increases as more disturbance-sensitive species (i.e., species at risk) are analyzed. We conclude that risk assessments that rely upon the most sensitive single species may substantially underestimate the risk of wildlife extinction across a planning area. Furthermore, the selection of a planning alternative based on relative threat of local extinction of wildlife populations can vary depending on which paradigm is used to estimate risk to viability across the planning area.

We review results and implications from recent wildlife studies that followed from the 1997 Tongass Land Management Plan (TLMP) and identify information needs and directions for research, development, and application. Sustained population viability of wildlife species was identified as a major issue in the TLMP planning process. Several species were identified as management indicator species, and research was conducted to determine their potential sensitivity to forest management. Southeastern Alaska was found to be a region with an especially high degree of endemism in its small mammal fauna, principally because of the combination of its archipelago geography combined with highly dynamic glacial history. Two species of endemic small mammals selected for demographic study, however, appeared to be less dependent on old-growth forests than had been suspected at the time TLMP was written: the northern flying squirrel ( Glaucomys sabrinus ) because of relatively high suitability of noncommercial, low-volume, mixed-conifer forest; and the southern red-backed vole ( Clethrionomys gapperi ) because of relatively high suitability of precommercially thinned young-growth forest. The northern goshawk ( Accipiter gentilis ) was found to be problematic for “management indicator” status because of logistical difficulties involved in monitoring this relatively rare, highly mobile species that frequently changes nest sites. Sampling protocols were developed for marbled murrelet ( Brachyramphus marmoratus ), although murrelet populations do not appear to be in trouble on the Tongass. The conservation strategy of TLMP for American marten ( Martes americana ) appeared to be sound on Chichagof Island where marten have been studied intensively, but implications for the rest of southeastern Alaska were unclear without further work. Studies of the Alexander Archipelago wolf ( Canis lupus ligoni ) indicated that population density of black-tailed deer ( Odocoileus hemionus ) and road access (to wolf hunters) were the predominant factors affecting wolf productivity and mortality, respectively. Finally, studies of bird community response to timber-harvest alternatives to clearcutting indicated that although creation of forest “edge” may increase nest predation rates, the actual response depends on a broad array of factors and is highly variable. We suggest that research, development, and application focus on plant and animal communities and management of vegetation to achieve specific objectives for wildlife habitat. We suggest that such efforts emphasize silviculture of second-growth forests, understanding old-growth reserves, distribution of endemic small mammals, and alternatives to clearcutting. Models for evaluating black-tailed deer habitat and populations are needed for subsistence-hunting management, and some work needs to be directed at interactions between tourism and selected wildlife species.

The dynamic geological history and naturally fragmented landscapes of southeastern Alaska create an environment with a high potential for endemism. The temperate rainforest of the region regenerates and develops slowly, and old-forest characteristics do not appear until ≥300 years following disturbance. The challenges of managing forest resources are intensified in island archipelagos because of the increased sensitivity of indigenous biota to disturbance and higher rates of extinction, especially among endemic organisms. Early expeditions of the large islands of the Alexander Archipelago (≈1% of all named islands) documented 27 endemic mammalian taxa. More recent studies with modern techniques found that some reputed endemics showed nominal levels of genetic divergence from other conspecific populations, but more divergence existed among several taxa than was reflected in the current taxonomy. Furthermore, the mammal fauna of southeastern Alaska has a nested structure with complex phylogeographic patterns suggesting multiple colonization events. Of eight taxa examined through phylogeographic analyses, five species showed acute genetic variation and divergence in mitochondrial sequences. Four species were comprised of coastal and continental clades (i.e. populations of recent common descent), whereas the fifth species showed a third clade. Conversely, the northern flying squirrel (Glaucomys sabrinus) and southern red-backed vole (Clethrionomys gapperi) were represented by relatively shallow divergent lineages. Still, G. sabrinus showed a distinct mitochondrial lineage on 11 islands (Prince of Wales Island complex), which exhibited severely reduced genetic variation. Moreover, flying squirrels in southeastern Alaska are genetically distinct from populations in the Pacific Northwest; as different as each is from the southern flying squirrel (G. volans). Recent ecological studies of endemic populations of the northern flying squirrel and the southern red-backed vole (reputed old-growth associates) suggest that risk of extirpation in managed landscapes is likely less than was presumed during recent land management planning in the region because: (1) abundant noncommercial forests apparently contribute to breeding populations of northern flying squirrels, which appear to have a more general lifestyle than populations in the Pacific Northwest; and (2) red-backed vole populations may be able to exist in managed young-growth stands that originated from clearcut logging. Still, there are essential questions for both species regarding the influence of annual population fluctuations on habitat distribution and population demography, stand and landscape features that restrict dispersal, and vegetative and structural characteristics of second-growth stands that will sustain breeding populations of both species.

American black bear (Ursus americanus) populations were significantly reduced throughout their range, particularly in southeastern North America. Currently, populations in this region are very fragmented, resulting in concern over possible barrier effects of rivers to normal bear movements and dispersal. This is particularly true for Mississippi, where black bear dispersal into the state is critical if populations are to be recovered. Thus, we studied the relative effects of rivercourses on bear movements and dispersal patterns in southeastern Arkansas, 1992–1996. We captured, radiocollared, and uniquely tagged 40 bears and used radiotelemetry to determine their movements. The Mississippi River (width ≈1600 m) deflected bear movements, whereas the White River (width ≈200 m) was not a barrier to bear movements or dispersal patterns. Frequency of river crossing differed by gender (P=0.007) and season (P

We examined the effects of white-tailed deer (Odocoileus virginianus) browsing and canopy opening size on relative abundance and diversity of woody and herbaceous regeneration in various sized forest openings in a southern bottomland hardwood forest over three growing seasons (1995–1997). We created 36 canopy openings (gaps), ranging from 7 to 40 m in radius, by group selection timber harvest in December 1994. Fenced exclosures were constructed in the center of each gap and vegetation was sampled monthly from April to September. Plant species richness, diversity, evenness, relative abundance, and a browsing index were calculated for each gap size and for each exclosure type. Herbaceous richness, diversity, or evenness did not differ among exclosure types in any year of the study. Browsing index was higher in the controls in 1996 and 1997. Browsing index for woody species was highest in the controls in 1995 and 1997. Relative abundance of herbaceous species was highest in the 29 m gap size in 1997. Richness and diversity of woody species were lowest in the 29 m gap size in 1995 and 1996. Overall browsing rates on both woody and herbaceous vegetation were low throughout all the 3 years of the study. Low browsing rates reflect seasonal changes in habitat use by deer. Because of the low rates of browsing, vegetative differences among exclosure treatments and gap sizes likely are not attributable to deer herbivory. Other factors, such as soil disturbance, may have influenced the initial vegetative response more than herbivory or gap size.

The density of active nests of bald eagles Haliaeetus leucocephalus decreased with increasing proximity to clearcuts; nesting success (number of chicks/nest) did not vary with density and proximity to clearcuts. Our results indicate that buffer zones around eagle nests probably should be at least 300 m wide to maintain eagle nesting density. Still wider buffers might be needed to include adequate alternative nest sites and perch trees, but the optimal width of the buffer strip may vary with other locally important factors, such as food availability.