Land-use and land-cover change in a highly fragmented boreal forest: The Beaverhills case study.

摘要

A structured hierarchical satellite imagery classification was applied to Landsat Multi Spectral Scanner and Thematic Mapper images from 1977, 1987 and 1998 in order to analyze land-cover and land-cover change in the Beaverhills region of central Alberta, Canada. This method allows for land-cover field data to guide the classification process, resulting in the optimum separation of spectral classes matching the land-cover classes of the field data.;The results suggest that the region contains significant amounts of forest and wetland habitat that are isolated by agricultural (crop and permanent forage) land use practices. The rate of deforestation in this region is similar to the broad trend at the southern periphery of the Canadian boreal forest region. The annual rate of change in forest cover was -0.82%/yr in the surrounding lands, and +0.61%/yr in the Beaverhills. Potential exists in the Beaverhills for the creation of a network of protected areas based on the UNESCO-MAB biosphere model.