ORIGINAL PAPER Conservation planning in a subdivided world

Abstract The identiWcation of priority areas for conservation tends to take place over two fundamentally diVerent spatial extents. First, there are analyses conducted at global or large biogeographic extents. Second, there are those conducted within geopolitical units. In this paper we show, using data for North American mammals, that spatial extent can have a profound eVect both on the number and locations of the priority areas identiWed to attain a particular conservation goal. For example, applying the same selection target to obtaining just a single representation of each species, the numbers of areas required increased by approximately an order of magnitude between treating North America as a single unit and treating the provinces separately. Although this scenario is undoubtedly extremely simplis-tic, such large diVerences are maintained with greater occurrence targets. Balancing the beneWts and disadvantages of conservation planning at diVerent spatial extents is not straightforward. However, a multi-scale approach that exploits the respective beneWts and downplays the disadvantages when focussing on smaller or larger extents would seem valuable