National‐scale conservation assessments at an appropriate resolution

Abstract

Abstract. Most national‐scale conservation assessments are carried out at a resolution that is different from the actual size of protected areas in the study region. Coincidence between nature reserves and both hotspots (areas of high species richness) and complementary areas (sets of sites within which all species are represented) have been reported. However, the resolution (size of grid cells) of the species’ distribution data upon which many of these studies are based is often close to an order of magnitude larger than the size of the reserves. Presumably, only a proportion of the species recorded in the coarse grid cells will actually be present on reserves. We use fine (2 × 2‐km square grid cells) and coarse (10 × 10‐km square grid cells) resolution data of national distributions for breeding birds throughout Great Britain, and presence data for avian species on Royal Society for the Protection of Birds (RSPB) nature reserves, to investigate the proportion of species in the local area (100 km2) that are actually present on reserves. RSPB reserves contain between 50% and 70% of species from the local area. These proportions are significantly higher than for randomly selected, non‐reserve areas, indicating that RSPB reserves contain higher concentrations of bird species than the wider countryside. Furthermore, on RSPB reserves these proportions of threatened and non‐threatened species are equal, whereas in non‐reserve areas the proportions of non‐threatened species are significantly higher than threatened species. Thus, reserves hold a higher proportion of threatened species than occurs in the wider countryside.

Knowing But Not Doing: Selecting Priority Conservation Areas and the Research–Implementation Gap

“…It is possible that our results overestimate the extent of this gap because authors may be unaware of implementation activities that use their research. Nevertheless, we regard this as unlikely, because practitioners typically do not access the peerreviewed literature (Redford & Taber 2000) in search of techniques to implement, and most implementing organizations have developed their own (often unpublished) conservation assessment techniques (Prendergast et al 1999;Hopkinson et al 2000). In addition, the researchimplementation gap may be narrower in the grey literature, which we did not analyze because it is not systematically accessible.…”

“…Rather than doing conservation, researchers appear preoccupied with describing the lack of representativeness of existing protected-area networks, experimentally testing data, and improving the efficiency of area selection algorithms in theory (Rodrigues et al 2000;Knight et al 2006a). The activities of conservation organizations rarely appear to be informed by published research (Pullin et al 2004), and conservation and land management organizations typically develop their own conservation assessment techniques independently of research in published journals (Prendergast et al 1999;Hopkinson et al 2000).…”

Vascular plant and bryophytes species representation in the protected areas network on the national scale

“…Gap analysis makes it possible to determine the situation of actual protected areas against the potential situation and notice differences between them (Rodrigues et al 2004). Unfortunately, often only species occurrence data in different size of grids are available instead of data for whole protected areas (Hopkinson et al 2000). In our study, the attention is focused on protected areas and IPAs with species lists.…”

“…The fact that specially selected areas for birds are richer in species than randomly selected areas (Hopkinson et al 2000) permits one to believe that the same could be applied to other species groups. It is possible to use a single species group to assess the network of protected areas but the result may not be true for other species groups.…”

Using climatically based random forests to downscale coarse-grained potential natural vegetation maps in tropical Mexico

“…There is a general concern that the resolution of traditional maps of potential vegetation such as that based on Rzedowski's classification are too coarse for real-world applications (Stoms 1992;Bredenkamp et al 1998;Hopkinson et al 2000;Kunin et al 2000;Hulme 2003;Rouget 2003;Hartley et al 2004). Map inaccuracies can result from the scarcity of information on natural and nearnatural vegetation (Zerbe 1998), and from the coarse resolution in the available maps of predictor variables (van Etten 1998).…”