|dc.description.abstract||Forest fragmentation is a continuing global concern. Human development destroys forest habitat, which separates and isolates forest fragments, decreases animal movement between fragments, reduces animal abundance and alters community composition. Forests are also becoming increasingly fragmented by linear forest clearings, such as roads, which are widespread throughout forests in tropical, temperate and boreal forest biomes. The ecological effects of roads have been well documented, but forests also contain extensive networks of non-road linear forest clearings (linear clearings hereafter). For example, forests overlying petroleum reserves may contain seismic lines, a common, but relatively narrow type of linear clearing used in sub-surface exploration. Seismic lines remain open with little tree growth for decades following their initial development, which alters abiotic conditions and plant communities inside seismic lines and at their edges. Therefore, while seismic lines contribute to forest fragmentation, little is known about their relative importance, compared with more common types of linear clearings. In addition, previous research has identified a number of fine-scale responses by birds to linear clearings, but less is known about other vertebrate taxa, including mammals. Furthermore, few studies have assessed their effects on wildlife communities.
This study investigated the ecological effects of seismic linear clearings, in a region of boreal forest in western Canada, where petroleum exploration and development has left a network of seismic lines over the past several decades. The first major aim of this study was to quantify the amount of forest fragmentation by seismic lines, compared with more well-known types of linear clearings. Second, this study investigated seismic line effects at the fine scale, on movement and habitat usage, by a community of mammals which varied widely in body-size and feeding type. Third, this study investigated whether mammal preferences for seismic line or forest habitat could be predicted by the environmental conditions in surrounding forest. Finally, it assessed effects at the landscape scale, by investigating how fragmentation by seismic lines affected the relative abundances and community composition of terrestrial mammals.
Forest fragmentation was quantified with a geographical information system (GIS) analysis, which used existing land cover maps and multiple sample plots of various spatial extents (5 ha to 4900 ha) relevant to mammals expected to exist within the study area. The investigations of seismic line effects on mammals were conducted in 14 study sites, dispersed across the study area. Mammal sampling occurred over three years and involved nine winter snow track surveys of two transects (1 km each), which were established on each site (one seismic line transect and one forest transect). Movement and local habitat use were assessed by comparing track measures, derived from seismic line and forest transects. The effects of environmental condition on mammal preferences for seismic line or forest habitat were assessed using measures of environmental conditions in the landscapes surrounding seismic lines, and these same landscape variables were used to assess effects of fragmentation by seismic lines on the relative abundances and community composition of terrestrial mammals.
Seismic lines dissected forest more than other types of linear clearings. They directly occupied a small land area, but seismic lines accounted for a large proportion of all edges (80%). Their density was double that of all other types of linear clearings (roads, rail, powerlines and pipelines), combined. Although the effects of seismic lines on mammals were assessed separately, integration of the results from this study revealed novel information about the seventeen mammal taxa (species or groups of similar species) and five functional groups which were recorded. Seismic lines affected mammals differently, depending on body size, food chain hierarchy (predator versus prey) and the scale of effect (fine versus landscape). Large mammals were most affected by seismic lines, which facilitated their movement throughout the forest. Large predators and large herbivores both moved along seismic lines. All of the large predators and the largest herbivore group (moose / elk) preferred seismic line habitat over forest habitat. Predator preferences for seismic lines were strongest in landscapes with the most continuous forest cover and flat terrain, while the largest herbivores (moose / elk group) also preferred seismic lines in landscapes with the most continuous forest cover. However, fine-scale effects translated consistently to the landscape scale for only one large predator (gray wolf), which was relatively more abundant in the most fragmented landscapes. The largest herbivore (moose / elk group) reacted oppositely at the landscape scale and was relatively less abundant in the most fragmented landscapes. None of the other large mammals strongly responded to seismic lines at the landscape scale.
Seismic line effects decreased as body size decreased. Some mammals with intermediate body sizes (e.g. marten) responded to seismic lines at the fine scale, but none responded at the landscape scale. Mid-sized predators moved along seismic lines, but both mid-sized predators and herbivores preferred forest habitat over seismic line habitat. Mid-sized predators preferred seismic lines in landscapes with flat terrain, despite their overall preference for forest habitat in the study area. Mid-sized herbivores preferred forest most strongly in flat terrain. Abundances of mid-sized predators and herbivores were unaffected by the level of forest dissection by seismic lines, at the landscape scale.
Small mammals (i.e. vole, mouse, shrew) were unaffected by seismic lines. This study found no evidence that any of the small mammal taxa, or the broader functional group, were strongly restricted in their movements, since all crossed seismic lines at rates similar to their rates of forest crossing. This finding is contrary to the expectation that seismic lines would restrict the smallest and least mobile mammals. Similarly, there was no evidence that small mammals preferred or avoided seismic lines, or their edges, compared with forest habitat, and there was no evidence that small mammal abundance was affected in landscapes fragmented by seismic lines.
This study of the effects of seismic lines on mammals in a boreal forest region revealed patterns of habitat- and landscape-scale effects which were substantially different from previous research into the effects of large scale clearings. Seismic lines did not result in an appreciable decrease in the overall amount of forest habitat. They also did not increase isolation among the remaining forest patches. Instead mammals regularly crossed seismic lines from one forest patch to another, and seismic lines facilitated movement, especially by the largest mammals inhabiting the study area. Forest fragmentation by seismic lines was associated with a reduction in the relative abundance of the largest herbivores in the study area and an increase in relative abundances of one of the large predators. These findings highlight the need for caution when generalizing fragmentation effects toward mammals inhabiting forests which contain seismic lines and other linear clearings. Clearings such as seismic lines, which dissected the forest, without increasing isolation among forest patches, were associated with much different effects than larger clearings. Furthermore, mammal preferences, for seismic line or forest habitat, often did not lead to landscape scale effects and these preferences were modified by environmental conditions in the surrounding landscapes. Depending on management objectives, existing seismic lines may require actions to reduce their effects, especially on large mammals, and new seismic lines should avoid landscapes where seismic line effects are likely to be strongest.
Further research is needed for fine-scale and landscape-scale effects in forest biomes outside boreal forest and community level research is needed during the snow free period in the boreal forest region.||