Premium
Conversión a Gran Escala del Bosque a Agricultura en las Llanuras Boreales de Saskatchewan
Author(s) -
Hobson Keith A.,
Bayne Erin M.,
Van Wilgenburg Steve L.
Publication year - 2002
Publication title -
conservation biology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.2
H-Index - 222
eISSN - 1523-1739
pISSN - 0888-8892
DOI - 10.1046/j.1523-1739.2002.01199.x
Subject(s) - deforestation (computer science) , geography , land cover , physical geography , boreal , biodiversity , land use , forestry , taiga , thematic mapper , agriculture , agricultural land , agroforestry , ecology , environmental science , satellite imagery , remote sensing , archaeology , computer science , biology , programming language
Despite widespread recognition of the importance of forest loss and fragmentation on biodiversity, the extent and rate of forest loss even in temperate regions remains poorly understood. We documented forest loss and assessed whether road density, rural developments, land quality for agriculture, and land ownership influenced the distribution and rate of change in forest cover for the entire boreal transition zone (49,846 km 2 ) of Saskatchewan, Canada. We used landscape data taken from the Canadian Land Inventory database in forest cover (1996) and Landsat thematic mapper data (1994) to study changes between 1966 and 1994. Forest covered 17,873 km 2 of the study area in 1966 and 13,504 km 2 in 1994. This represents an overall conversion of 73% of the boreal transition zone in Saskatchewan to agriculture since European settlement and an annual deforestation rate of 0.89% over the last 28 years, a rate approximately three times the world average. Mixed‐regressive, spatially autoregressive models explained a considerable portion of the variation in forest cover ( r 2 = 0.83) remaining on the landscape and suggested that less forest remained on lands that (1) were privately owned, (2) had soils with high suitability for agriculture, (3) had high road density, and (4) were in the southern portions of the study area. Strong spatial autocorrelation in the data indicated that areas of remaining forest tended to be spatially clustered. Our ability to predict where deforestation occurred between 1966 and 1994 was poor when we excluded the spatial autocorrelation terms from our model, but it was clear that deforestation was more likely to occur on privately owned lands than on those managed by the provincial government. Despite dramatic changes to forested areas in the boreal transition zone, and despite the importance of this area to a wide variety of forest‐dwelling wildlife, no programs are in place to slow or halt deforestation.