Date of Award

5-2009

Level of Access Assigned by Author

Campus-Only Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Wildlife Ecology and Wildlife Conservation

Advisor

Daniel J. Harrison

Second Committee Member

William B. Krohn

Third Committee Member

Steven A. Sader

Abstract

The American marten (Martes americand) prefers habitats with complex physical structure associated with mature, closed-canopy forest, which provides protection from predators, resting sites, and access to prey. Previous research has concluded that timber harvesting can negatively influence marten density, and clearcut harvesting has been implicated in local population declines. Studies also suggest that martens may be particularly sensitive to habitat loss, predicting a steep decline in probability of home range occupancy in response to a small decline in percent suitable habitat. Little is known, however, about the cumulative effects of habitat loss and fragmentation on habitat supply for martens or the spatiotemporal dynamics of probability of occurrence for martens in a managed landscape. I developed a spatially-explicit time series of marten habitat, derived from satellite imagery, to evaluate effects of forest management on quantity and distribution of marten habitat, on spatiotemporal patterns of marten occurrence, and on estimated densities of martens in a dynamic forest landscape 1975-2007. Timber harvesting was widespread during this period and habitat that previous research has defined as suitable by martens declined by 434,978 ha (32%) as a result of stand-replacing harvests 1975-2007. Declines in probability of occurrence followed two spatiotemporal trends. The majority of loss occurred in the first 16 years of the time series (1975-1991), resulting from salvage logging that occurred in response to the 1973-1985 spruce budworm (Choristoneura fumiferana) outbreak. Rapid declines in probability of occurrence for both male and female martens occurred where large blocks of mature spruce-fir forest were salvaged. Salvage logging also contributed to fragmentation of marten habitat with a 5.6-fold increase in the number of suitable habitat patches 1975-1991. Declines in habitat supply and probability of occurrence continued 1991-2007, coinciding with the broad-scale changes in timber harvesting patterns, which resulted from the implementation of new forest policies that contributed to a reduction in clearcut harvesting but a widespread increase in partial harvesting and total acreage harvested. The cumulative effects of loss and fragmentation of marten habitat 1975-2007 negatively influenced the percent suitable habitat and habitat configuration at the scale of a marten home range. These changes resulted in widespread declines in probability of occurrence for males and female martens and a substantial reduction in the number of townships with a potential marten density of >1 martens/km2. Additionally, 307,862 ha (33%) of marten habitat received a partial harvest 1988-2007, which cumulatively affected >90% of potential marten home ranges. Partial harvesting may increase the spatial requirements of martens, suggesting that declines in potential marten densities are conservative. Further, based on previously published structural thresholds for marten habitat use, it is likely that some proportion of the areas that have received a partial harvest no longer represent suitable habitat for martens. Preliminary estimates suggest that in many partially-harvested stands the residual basal area is less than published thresholds for marten habitat use. This indicates that the actual loss of habitat 1975-2007 was between 32% and 54%, and strongly suggests that additional research is needed to determine the extent that partial harvesting is affecting habitat for martens, and potentially compounding the effects of habitat loss and fragmentation that are reported herein.

Download

Files over 10MB may be slow to open. For best results, right-click and select "save as..."

Share