Walt Klenner. 1998. Changing Landscapes in Smith, I.M., and G.G.E. Scudder, eds. Assessment of species diversity in the Montane Cordillera Ecozone. Burlington: Ecological Monitoring and Assessment Network, 1998.

CHANGING LANDSCAPES

Monitoring forested habitats in the Montane Cordillera Ecozone
across large spatial and temporal scales

Walt Klenner

2. PATCH SIZE MOSAIC OF FOREST AND OPENINGS

Historically, a fine-scale interspersion of seral stages that created abundant edge habitat was thought to favor wildlife (Leopold 1933). More recently however, the ecological value of small habitat patches that are primarily edge habitat has been examined more closely and concerns have been raised over the fragmentation of habitat (Harris 1984, Wilcox and Murphy 1985, Wilcove et al. 1986, Hunter 1990, Andren 1994). Habitat fragmentation in forests involves changes in both habitat structure and pattern, including: (i) a decrease in the amount of late seral habitat, (ii) a shift in the size distribution of habitat patches, with fewer large patches and a higher proportion of small patches (e.g. less than 40 ha) than existed in the unmanaged landscape, with an associated loss of forest interior conditions, (iii) the loss of connectivity between old forest habitats, and (iv) usually an increase in the number of roads that bisect a landscape.

The landscape patterns necessary to maintain species and ecological processes are unclear. Studies in areas where forests are surrounded by agricultural or suburban lands suggests a positive relationship between increasing patch area and species persistence and abundance (MacClintock et al. 1977, Urban et al 1987). This pattern is less clear where forestry is the main land use practice and where the landscape is a dynamic and changing mosaic of early, mid and late successional forest (McGarigal and McComb 1995, Schiek et al. 1995). Large patches (e.g. greater than 200 ha) of old forest are important to some species because of the special microclimates these habitats provide (Harris 1984, Chen et al. 1990, 1993), or the large spatial requirements of some species (Harris 1984, Shafer 1990, Angelstam 1992, Essen et al. 1992).

Conventional forest management practices usually lead to a reduction in the abundance of large patches of old forest and an increase in the number of small patches (Harris 1984, Mladenoff et al. 1993, Klenner et al. 1995). For example, Spies et al. (1994) report that in a managed landscape in Oregon, less than 12% of private lands had patches of coniferous forest with forest interior conditions defined as greater than 100m from adjacent early seral habitat, compared to approximately 43% on public lands. Much of this difference was attributed to the dispersed cutblock pattern of harvesting and higher rates of harvest on many private lands. The only patches of greater than 1000 ha of contiguous interior forest were found on lands managed as wilderness areas or research natural areas.

Changes in the patch size distribution of forests in the MCE will need to be reported by broad age class categories, and be conducted at a scale that excludes excessive detail but captures some of the inherent differences between ecosystems. Reporting the patch size distribution within specific age categories (e.g. 0-40 yrs., 41- 80, 81-140 and 140+) and by major habitat associations within a biogeoclimatic zone will help standardize results and simplify categories. For example, habitat information already captured in inventories (e.g. inventory type groups that represent categories of stands with similar leading tree species composition such as Douglas-fir (Pseudotsuga menziesii) or lodgepole pine (Pinus contorta) would be suitable for establishing broad habitat categories.

Tree species composition.

Forest management practices that favor the early dominance of a site by conifers, or shift the species composition of the stand to fast-growing conifers such as lodgepole pine pose a risk to maintaining biological diversity if practiced on an extensive scale. There is general agreement that maintaining a component of broadleaved trees in coniferous forests increases the species diversity of birds and invertebrates. Birds generally choose different vegetation types or strata for foraging or nesting, hence more complex vegetation structures, critical habitat attributes such as snags or mixed species stands will likely support a greater diversity of birds (Morgan and Wetmore 1986). Huff and Raley (1991) examined 132 Douglas-fir stands and concluded that even small inclusions of hardwoods increased bird species diversity. The association with broadleafed trees is clear for cavity-nesting birds which often prefer these trees for excavation (Kiesker 1987, Raphael 1987) since the decay of heartwood usually begins earlier than in conifers. Bunnell et al. (1991) estimated that between 8.5 and 11% of the bird fauna in British Columbia is strongly associated with broadleaved trees. Several mammals also prefer hardwoods for some life requisites. Fisher (Martes pennanti) prefer black cottonwood (Populus trichocarpa) and aspen (P. tremuloides) as denning sites (Weir 1995), and some small mammals are more common in hardwood stands in coastal forests (McComb et al. 1993 a,b, Hagar 1990).

Planting selected tree species and vegetation management practices to control competing broadleaf vegetation are two activities that may affect succession and alter tree species composition. Harding (1994) noted that of the tree species planted in BC in 1989-90, the vast majority were either Englemann or white spruce, lodgepole pine or Douglas-fir. These practices may diminish not only the abundance of broadleafed species, but also conifers such as western hemlock (Tsuga heterophylla) that are not favored in planting or stand tending activities. Whether the planted species continue to dominate the site on a long-term basis without further intervention is not clear, but should be monitored.

Density and dispersion of roads.

Conventional dispersed cutblock harvesting requires an extensive and permanent network of roads to access timber on an ongoing basis. Roads can have negative impacts on an ecosystem by: 1. forming barriers to dispersing organisms, 2. providing corridors along which invading organisms (e.g. weeds) enter an ecosystem, 3. increasing the incidence of human-caused fires, 4. facilitating excessive legal hunting and/or poaching, and 5. contributing to soil erosion and sedimentation of aquatic habitats in steep terrain or in areas of unstable soils (Thomas et al. 1976, Mader 1984, McLellan and Shackleton 1992, Thurber et al. 1994,).

Several studies have described patterns of increasing landscape dissection caused by roads (Hummel and Pettigrew 1991, Harding and McCullum 1994b, Miller et al. 1996, Reed et al. 1996, Northcote, this volume), but few have examined the impacts of roads on biological diversity. Direct traffic-caused mortality has been reported by several researchers. Bellis and Graves (1971) noted that white-tailed deer fed on grassy shoulders and were consequently hit by cars. Oxley et al. (1974) and Adams and Geis (1983) document mortality of small mammals during road crossings, but noted that that mortality did not seem to affect population viability. Roads facilitate increased access by hunters and poachers to large carnivores and ungulates (McLellean 1990) and may compromise the viability of some populations. Roads may serve as barriers to the movements of some animals (Oxley et al. 1974, Schreiber and Graves 1977). Wider roads are usually more effective as barriers, and roads wider than 20m form an effective barrier to small mammals. These studies demonstrate the diverse impacts roads have on ecosystems, and suggest that roads can serve as critical indicators of past or likely future development.

Throughout the MCE, most roads are built to facilitate resource extraction activities such as forestry and mining, and are subsequently used for recreation. Vold (1992) classified land in BC with respect to road access and patch size. Primitive areas were characterized as being greater than 5000 ha and at least eight km from an access road. Only about 25% of BC fell into the primitive category, and almost all of this occurs in steep mountainous terrain in either the Boreal Cordillera or Pacific Maritime Ecozones. Most of the forested land in the MCE, especially habitats at lower elevations (e.g. the Interior Douglas-fir biogeoclimatic zone) is accessible by two or four-wheel drive vehicles. Other habitats in the MCE have fewer roads (e.g. the ESSF and MH zones), but roads and access should be monitored throughout the MCE and reported in relation to biogeoclimatic zones or subzones to establish trends.

Landbase status.

Parks and wilderness areas that are protected from natural resource extraction provide refugia for species sensitive to natural or managed disturbances, and serve as baselines against which other areas can be compared (Noss 1983, Morrison and Turner 1994, Arcese and Sinclair 1997). Land use and ownership changes often occur gradually, but as public lands are converted to private holdings, forested lands to agriculture (Mladenoff et al. 1993), or agricultural lands to urban developments, significant cumulative ecological changes occur. Changes in the status of public lands, such as the loss or acquisition of parks and protected areas (Morrison and Turner 1994), are important management actions that will affect our ability to manage for biological diversity. Monitoring ownership or landbase status trends over time can serve as an early warning system to alert managers of the scarcity or loss of certain habitats, especially when the rate of habitat change is slow and ecological impacts appear as long-term cumulative effects.