Identification of de facto protected areas in boreal Canada
*Subscription may be required
Canada is dominated by remote wilderness areas that make important conservation contributions, but are currently only protected de facto by their inaccessibility. Mechanisms for the identification and formal protection of such areas can help ensure that they continue to function naturally and provide essential ecosystem services. However, a lack of spatially explicit, publicly available sources of data on anthropogenic disturbances and natural resource extraction challenges the development of detailed wilderness inventories. We suggest that landscape structure can be used to classify areas of natural landscapes, as trained by the landscape structure of protected areas, and demonstrate this approach by mapping de facto protected areas in Canada's boreal forest. Overall, between 50%, based on landscape structure, and 80%, based on anthropogenic infrastructure alone, of Canada's boreal zone exists in large, intact blocks. The true extent of boreal wilderness likely falls within this range, as existing infrastructure datasets may omit disturbance and the protected area network in far northern areas proved inadequate to train effective wilderness classifications. We anticipate that such efforts may be improved by refining the identification of training areas or by classifying along additional landscape metrics. Nevertheless, the areas identified are valuable candidates for protected area expansion, and can contribute to a reserve network that meets national and regional conservation targets and is representative of the range of vegetation productivities, which was used as a biodiversity surrogate. Our general approach need not be limited to the boreal forest, as it has the potential to successfully identify relatively undisturbed (or less disturbed) areas over a range of systems and across levels of human influence.
|Publication Type:||Journal Article|
|Copyright:||© 2011 Elsevier B.V.|
|Item Control Page|
Downloads per month over past year