SEARCH

SEARCH BY CITATION

Keywords:

  • effectiveness;
  • Extent of Occurrence;
  • Important Bird Area;
  • IUCN Red List;
  • site-based conservation

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Protected Areas (PAs) form a core component of efforts to conserve biodiversity, but are designated for a variety of reasons. We assessed the effectiveness of PAs in covering the ranges of 157 globally threatened terrestrial bird species in mainland Africa and Madagascar. To reduce commission errors, rather than using Extent of Occurrence (EOO) as a measure of distribution, we estimated the Extent of potentially Suitable Habitat (ESH) for each species within its EOO, using data on habitat preferences and land cover. On average, 14% of species' ESH fell within PAs, with negligible coverage of Critically Endangered species. By contrast, an average of 30% of species' ESH fell within Important Bird Areas (IBAs), a network of sites identified using globally standardized criteria as critical for bird conservation. IBAs that overlapped or fell within PAs were significantly less effective at covering the ESH of threatened birds than those falling outside the PA network, and for IBAs partly overlapping with PAs, coverage of threatened birds was significantly greater in the unprotected part. Expansion of the PA (and IBA) networks in parts of Madagascar, the Albertine Rift, Cameroon Highlands, Eastern Arc and eastern Kenya would benefit globally threatened bird species conservation.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Site-based conservation is widely recognized as being important in tackling global biodiversity loss (Joppa, Loarie & Pimm, 2008; Nagendra, 2008; Mwangi et al., in press), and is considered to be the highest single immediate priority for the conservation of 82% of globally threatened mammals, birds, amphibians, tortoises and turtles (Boyd et al., 2008). Protected Areas (PAs) are sites protecting biodiversity, natural resources and associated cultural resources, managed through legal or other effective means (Dudley, 2008). Some 2.2 million km2 (7%) of Africa is included within PAs (UNEP-WCMC, 2007), but this is less than the global average of 12.9–13.4% (Jenkins & Joppa, 2009; Coad et al., 2010). PAs are often sited opportunistically or targeted at charismatic and financially important megafauna, resulting in an inefficient representation of species and habitats within PA networks (Pressey et al., 1993).

In response, there have been a number of attempts to identify, using standardized criteria, sites of conservation importance that form cohesive and complementary networks (Brooks et al., 2006). Among these is the Important Bird Area (IBA) network (BirdLife International, 2008a). The significance of IBAs for prioritizing conservation efforts is widely recognized; many governments have used IBA inventories to inform PA network expansion (BirdLife International, 2008a), and the degree of coverage of IBAs by PAs is used to measure progress in tackling biodiversity loss (Butchart et al., 2010; SCBD, 2010) and for tracking sustainable development towards the UN Millennium Development Goals (UN, 2010). IBAs are identified nationally, through multi-stakeholder processes, using objective and quantitative criteria relating to globally threatened, restricted-range, biome-restricted and congregatory bird populations (BirdLife International, 2008a). African IBAs cover 2.1 million km2 (Fishpool & Evans, 2001), an area comparable to the extent of African PAs. Two-thirds of African IBAs support significant populations of globally threatened species (Fishpool & Evans, 2001).

An analysis of the extent to which the ranges of globally threatened species are covered by PAs [based on Extent of Occurrence (EOO) and PA overlap] suggests that coverage is far from complete (Rodrigues et al., 2004b), especially because analyses using EOO tend to overestimate the likelihood of a species occurring at any particular site (Rondinini, Stuart & Boitani, 2005; Rondinini, Chiozza & Boitani, 2006a; Jetz, Sekercioglu & Watson, 2008). Species do not usually occur throughout their entire EOO (Gaston & Fuller, 2009), but accurate mapping of the inhabited area within the EOO [Area of Occupancy (AOO)] requires detailed field surveys, which are expensive, time consuming and therefore lacking for the vast majority of species (Jetz et al., 2008). The maximum potential extent of the AOO can be estimated by combining data on altitudinal preferences, habitat use and land cover within EOOs to determine the Extent of Suitable Habitat (ESH). Approaches vary in precision, from species-specific models (e.g. Osborne, Alonso & Bryant, 2001; Donald et al., 2009) to the use of broad habitat associations for species and general land cover maps (e.g. Rondinini et al., 2005; Buchanan et al., 2008).

There has been no previous continental scale analysis of the efficacy of the PA network at covering the ESH of globally threatened species, even though this is a more realistic approach than assessing EOO coverage. Here, we explore patterns in the ESH of globally threatened African birds, assess the efficacy of the African PA network in overlapping the ESH of these species and compare this with the efficacy of the IBA network. As a result, we identify the areas of highest importance for future expansion of networks.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

We considered all 157 species of terrestrial birds breeding on mainland Africa or Madagascar that were classified by BirdLife International as globally threatened on the 2008 IUCN Red List, comprising nine Critically Endangered, 59 Endangered and 89 Vulnerable species (BirdLife International, 2008b; IUCN, 2008).

Data

Data on each species' altitudinal range and habitat associations were extracted from BirdLife's datasets (BirdLife International, 2008b), with a small number of gaps supplemented from the literature (del Hoyo, Elliot & Sargatal, 1992; Kear, 2005) and personal communication (R. Safford, L. Roxburgh, pers. comm. 2008). The ‘usual’ altitudinal limits for each species were used, excluding the upper and lower limits where a species may be ‘occasionally’ recorded. Altitudinal ranges were not available for 19 of the 157 globally threatened species (10 Vulnerable and nine Endangered species); in these cases, all suitable land cover within the EOO, irrespective of altitude, was included. The resultant percentage of the EOO that was considered ESH for these species did not differ significantly from that of the other species (t=0.23, P=0.81).

BirdLife codes the importance of each habitat used by each species. ‘Major’ habitats are those important for the survival of the species, either because it has an absolute requirement for the habitat at some point in its life cycle (e.g. for breeding or as a critical food source) or it is the primary habitat (or one of two primary habitats) within which the species usually occurs, or within which most individuals occur. ‘Suitable’ habitats are those in which the species occurs regularly or frequently. ‘Marginal’ habitats are those where the species occurs only irregularly or infrequently, or only a small proportion of individuals are found in the habitat. We included habitats under all three levels of importance to err on the side of overestimating, rather than underestimating, ESH. We used BirdLife International's digitized EOO maps for each species (based on published and unpublished literature and expert input) and digitized boundaries for IBAs. The latter were not available for 254 IBAs (23% of the total; see supporting information Table S3). The area excluded comprised 16.8% of the total IBA area, but two-thirds of this was accounted for by 10 IBAs, which fell in areas with no, or very few, globally threatened bird species (predominantly in the Sahara and the Horn). Overall, the IBAs for which digital boundaries were not available were smaller than other IBAs (median area=36.0 vs. 295.9 km2, W=104 555, P<0.001).

Altitude data were obtained from the USGS GTOPO30, which maps altitude at a spatial resolution of 30 arc seconds, or c. 1 km (USGS, 2008). Land cover assessment used the Land Cover Map of Africa 2000 (GLC2000; Mayaux et al., 2004), which maps the distribution of 27 major vegetation types across mainland Africa, Madagascar and surrounding islands, at a resolution of 1 km. The map was produced using Spot Vegetation satellite data (Mayaux et al., 2004). Habitat associations of species were coded using the IUCN habitats classification scheme (version 3.0: see http://www.iucnredlist.org/technical-documents/classification-schemes/habitats-classification-scheme-ver3). These were matched to the GLC2000 land cover classification, based largely on a generalized conversion but refined using additional details of specific land cover classes used by Mayaux et al. (2004) (see supporting information Table S1).

Boundaries of all PAs with IUCN classifications, Ramsar Convention Sites, World Heritage Sites, UNESCO-MAB Biosphere Reserves and sites within other international conventions and agreements represented by spatial boundaries, were obtained from the World Database on PAs (UNEP-WCMC, 2007). The use of this set of sites for PA analyses, which includes sites with international but not national designation, has been proposed by Coad et al. (2010) but criticized by Jenkins & Joppa (2010). Sites with no assigned IUCN category were not regarded as ‘protected’ for these analyses, as it is questionable whether the management of these areas provides effective protection for biodiversity (Bruner et al., 2001), and digital boundaries for such sites were often unavailable (UNEP-WCMC, 2007).

Analysis

ESH maps were created for each species by selecting the areas within its EOO and altitudinal range where land cover classes matched those used by the species. In ArcMap 9.1 (ESRI, 2006), maps of the PA and IBA networks were intersected with the ESH maps to estimate, for each species, their areas of overlap. All maps were projected into an equal area projection (Africa Sinusoidal). Thirteen species (all with EOO <20 000 km2) had ESH estimates of 0 km2, probably the result of errors in habitat coding, in the GLC2000 or altitudinal preferences datasets, or from their occurrence in habitat patches much smaller than 1 km2, which were therefore not represented on the map. These species were excluded from calculations of % ESH covered by sites. An accuracy assessment was carried out by comparing the presence and absence of species in IBAs predicted by the ESH analysis, with the independently recorded presence of species at sites from data held in BirdLife International's World Bird Database. These data only recorded species present in sufficiently high numbers to trigger IBA designation of sites, meaning that the analysis was conservative in terms of false presences (Fishpool & Evans, 2001). Direct comparison of coverage of the IBA and PA networks is valid despite their high degree of spatial overlap, as the two groups are independently defined. Throughout, all means are reported±se.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

EOO and ESH

On average, the ESH of a species comprised 27.6% (±2.20) of the area of its EOO, and although there was unsurprisingly a strong positive association between EOO and ESH, there was some scatter around this relationship (Fig. 1). There was a tendency for Critically Endangered species to have a smaller proportion of ESH within their EOOs than less threatened species (Fig. 2), but this difference was not statistically significant (F2,154=1.72, P=0.18). There was no difference in area of ESH between threat categories (F2,154=0.09, P=0.9), a reflection of the fact that range size is not the only criterion used in extinction risk categorization.

image

Figure 1.  Relationship between Extent of Occurrence (EOO) and Extent of Suitable Habitat (ESH) for 144 globally threatened terrestrial bird species in Africa (r2=0.710, P<0.001). The solid line shows the regression fit. The dotted line represents maximum possible ESH (i.e. ESH=EOO).

Download figure to PowerPoint

image

Figure 2.  Extent of Suitable Habitat (ESH) as a percentage (±se) of Extent of Occurrence (EOO) for globally threatened African birds by threat category. Numbers in brackets indicate sample sizes.

Download figure to PowerPoint

For the 157 species, 1003 ‘presences’ (a sufficient proportion of the global population to trigger recognition of the site) were recorded in the 866 IBAs. These data were compared with the overlaps between the ESH maps and the IBA polygons. Excluding those IBAs for which digitized polygons were missing, there were 2260 mismatches (Table 1). Of these, 1887 were ‘false presences’, where the ESH placed a species in an IBA in which independent data indicated that it did not occur in sufficient numbers to ‘trigger’ IBA criteria, and 373 were ‘false absences’, where a species triggered IBA identification, but the ESH analysis predicted its absence (including 29 occurrences for 13 species with ESH estimates of zero). Hence, there were five times more commission errors than omission errors. Of the false absences, 156 (41.8%) were a result of no overlap between digitized IBA and EOO boundaries, owing to errors in mapping the boundaries of EOOs (or, less likely, the location of IBAs). In the remaining 217 cases (58.2%), the IBA fell within the EOO, but not the ESH of the species, potentially due to errors in the habitat coding, land-cover map, habitat-land-cover matching, or the documented altitude ranges of the species. By way of comparison, if the analysis had been based on EOO rather than ESH, the number of false absences would have been reduced by 217, but false presences would have increased by 843. For individual species, the average concordance between recorded occurrences and IBA and ESH overlap was 98.21% (±0.39).

Table 1.   Comparison between the observed occurrence of species in Important Bird Areas (IBAs) and the overlap between IBAs and Extent of Suitable Habitat (ESH), and Extent of Occurrence (EOO) maps
 ESHEOO
Overlap with IBANo overlap with IBAOverlap with IBANo overlap with IBA
  1. Figures are for the total number of incidences of presences/absences for 157 globally threatened bird species across 866 IBAs.

Observed occurrence
 Presence630373847156
 Absence1887133 0722730132 229

Performance of PAs and IBAs

The absolute area of a species' ESH covered by sites increased with ESH area (rs=0.863, P<0.001 for PAs, rs=0.909, P<0.001 for IBAs; Fig. 3a and b). The percentage of ESH within PAs increased with increasing area of ESH (rs=0.255, P=0.002), but the percentage ESH within IBAs decreased with increasing area of ESH (rs=−0.235, P=0.005; Fig. 3c and d). This suggests that PAs provide better coverage than IBAs for threatened species with large ranges, while IBAs provide greater coverage than PAs for threatened species with restricted distributions.

image

Figure 3.  Relationship between Extent of Suitable Habitat (ESH) and (a) area of ESH within Protected Areas (PAs), (b) area of ESH within Important Bird Areas, (c) percentage of ESH within PAs and (d) percentage of ESH within PAs, for 144 globally threatened African bird species.

Download figure to PowerPoint

Of the 144 species considered, 37 had none of their ESH within PAs, and a further 85 had <25% of their ESH within PAs (see supporting information Table S2). By comparison, 14 species had no ESH within IBAs and a further 70 species had <25% of their ESH within IBAs. Thirty-one species were better covered by PAs than by IBAs, compared with 100 species better covered by IBAs than by PAs. While a full assessment of differences was precluded because some species are found in more than one land-cover type, it appeared that species associated with forest and woodland had a high percentage of ESH within both PAs and IBAs, while those associated with sparse vegetation were less well covered by PAs, especially when compared with their coverage by IBAs (Fig. 4).

image

Figure 4.  Mean percentage (±se) of Extent of Suitable Habitat (ESH) falling within Protected Areas (PAs) and Important Bird Areas (IBAs) for 144 globally threatened African bird species across six different land-cover types. Species may be present in more than one land-cover type and numbers in brackets indicate sample sizes.

Download figure to PowerPoint

On average, a significantly lower percentage of species' ESH fell within PAs (13.9%±1.7) than IBAs (30.4%±2.4) (Z=7.859, P<0.001). Average coverage by PAs was consistently lower than that provided by IBAs across all IUCN Red List threat categories (Z=2.023, P=0.043, for Critically Endangered species, Z=5.206, P<0.001, for Endangered species and Z=5.420, P<0.001, for Vulnerable species; Fig. 5). There was a significant difference in ESH coverage by PAs between threat categories, with less threatened species better covered than more threatened species (χ22=13.609, P=0.001; Fig. 5). Critically Endangered species, which are of the greatest conservation concern, were particularly poorly covered by the PA network, with an average of just 0.5% of ESH within PAs (compared with 29.2% for IBAs). The decline in PA coverage of species with increasing conservation concern exceeded that which might have been expected given the lack of difference in average ESH area across different threat categories. There was no significant difference in IBA coverage between threat categories (χ22=3.670, P=0.160). The contribution of IUCN Category V and VI PAs was similar to that which might be expected based on the area they cover, contributing 23.3% to the total area of the PA network, and on average, 19.1% to the area of threatened species' ESH within PAs. PAs designated specifically for the protection of biodiversity therefore did no better in covering the ESH of threatened birds than those designated for other reasons.

image

Figure 5.  Mean percentage (±se) of Extent of Suitable Habitat (ESH) for threatened African bird species contained within Protected Areas (PAs) and Important Bird Areas (IBAs). Numbers in brackets indicate sample sizes.

Download figure to PowerPoint

Within partially protected IBAs, 36 species appeared only to occur in parts of IBAs that were not protected, and for all species in such IBAs, the proportion of ESH that fell within the protected part was much lower than that in the unprotected part (8.8±1.51 vs. 21.6±2.5%; t=4.9, P<0.001). While some caution is needed in interpreting these results due to potential errors in the spatial data, unprotected parts of the IBA network appear to be of disproportionate importance for globally threatened birds.

Globally threatened birds are unevenly distributed across Africa; their species richness (based on ESH) ranges from zero across large parts of the continent, up to 10 species km−2 in hotspots such as the Ethiopian Highlands, South African plateau and Madagascan rainforests (Fig. 6a). The mean density of threatened birds was 0.74 (±0.0002) species km−2 outside PAs and IBAs, compared with 1.11 (±0.0002) species km−2 inside PAs [and 1.19 (±0.0002) species km−2 in IBAs]. Excluding 33 species with EOOs >100 000 km2, for which site-based conservation may be of lower relevance (Boyd et al., 2008), and taking a 25% threshold for coverage of ESH by sites, based on a step in the frequency distributions of coverage (supporting information Figure S1), PA coverage of threatened birds was the poorest in Madagascar, the Albertine Rift, Cameroon Highlands, Eastern Arc and eastern Kenya (Fig. 6b). IBA coverage was also the poorest in these areas (supporting information Figure S2), but both the extent of the poorly covered regions and the number of poorly covered species contained within them were considerably lower for the IBA network than for the PA network. Western Africa and the western coast of Madagascar were notable areas where PA coverage was poor, but IBA coverage good.

image

Figure 6.  Density of species based on Extent of Suitable Habitat (ESH) for (a) all globally threatened African bird species, mapped as number of species per km2 and (b) globally threatened African bird species with <25% of their ESH within Protected Areas, (mapped as maximum number of species per 50 km2 to increase clarity of figure, and excluding species with Extent of Occurrence >100 000 km2: see text).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

This study produced maps of ESH equating to maximum potential Areas of Occupancy for globally threatened birds in Africa, based on the distributions of suitable land cover at appropriate altitudes within their known EOO. It develops and complements previous habitat suitability mapping efforts carried out on African amphibians and mammals (Rondinini et al., 2005, 2006b; Rondinini & Boitani, 2006).

The approach was validated by comparing observed and suggested distributions of species in IBAs. Overall, the two agreed in 98.2% of cases. Therefore, in the absence of detailed AOO maps supported by field data, the approach of using habitat data to produce ESH maps is a useful means of refining distributional analyses. If unadjusted EOO maps had been used, the number of false presences would have been much greater, because the ESH of a species covered, on average, just 27.6% of the area of its EOO (values ranged from 0.0 to 94.5%). Previously, a comparison of EOO maps and field observations by Jetz et al. (2008) reported most bird species occurred in 40–70% of the range indicated by their EOO, with ‘proportional range overestimation’ between 0 and 91%.

Overlaying PA boundaries onto the ESH maps indicated that PAs covered, on average, just 14% of threatened species' ESH, and that this coverage declined sharply with increasing threat status. Owing to uncertainty over the effectiveness of their protection (Bruner et al., 2001), PAs with no IUCN category were excluded from this analysis and consequently, coverage may in fact be slightly higher than our results indicate. However, our results broadly concur with previous studies. De Klerk et al. (2004) reported that PAs in sub-Saharan Africa failed to cover over half of the threatened bird species in the region, and Rodrigues et al. (2004a,b), using EOO, reported that the global PA network failed to provide adequate coverage for a wide range of taxa, especially threatened species. By using the ESH for species, rather than EOO (Rodrigues et al., 2004a; Jetz et al., 2008), and quantifying the extent of coverage for individual species, our study provides stronger evidence for the inadequacy of the current PA network for the conservation of the most threatened species.

The proportion of a species' range that is required to fall within conservation areas for a network to be effective has not been defined (Rodrigues & Gaston, 2001), and IBAs are not intended to cover all (or even a high percentage) of threatened species' ranges (Fishpool & Evans, 2001). However, on average, 30.4% of the ESH of globally threatened species was covered by IBAs. This is likely to be an underestimate, as boundaries were not available for all IBAs. Also, 12 out of 13 species with an ESH estimate of 0 km2 occur in IBAs, and 13 out of 14 species with an apparently zero ESH within IBAs in fact have populations in at least one IBA – these results likely being errors from one or more sources (digital maps and/or species–habitat relationships). As such, IBAs are much more effective than PAs at covering ESH. The better performance of IBAs is in itself no surprise, because two-thirds of the IBAs considered were deliberately targeted to areas where threshold populations of globally threatened species occur, while PAs are designated for a wide variety of reasons in addition to conservation of threatened species, or even biodiversity in general. However, the poor performance of PAs compared with IBAs, and in particular the finding that PA coverage decreased with increasing extinction risk, is especially concerning considering the high level of overlap (61.6%) between IBAs and PAs in Africa.

This argues that the PA network in Africa should be expanded to improve coverage of species at the greatest risk of extinction, and that the IBA network provides a useful set of priority sites for achieving this. Within partially protected IBAs, the unprotected parts appeared to be of disproportionate importance for globally threatened birds, suggesting that extending existing PAs in these areas would be beneficial. This mirrors the situation in other regions: for example the European Commission has taken a number of countries to court for failing to designate sufficient PAs, citing IBA inventories as adequately robust datasets of priority sites upon which to base PA expansion (BirdLife International, 2008a). Similarly, the UN have recognized PA coverage of IBAs as a useful indicator of sustainable development towards the Millennium Development Goals (UN, 2010).

While targeting protection towards existing IBAs may be beneficial for most species, some were identified as being poorly covered by both the PA and IBA networks. After the exclusion of species with ranges >100 000 km2, such species were found to be concentrated in Madagascar, the Albertine Rift, Cameroon Highlands, Eastern Arc and eastern Kenya. Previous studies have also highlighted some of these regions as being of high conservation priority and requiring an increased conservation effort (Stattersfield et al., 1998; Médail & Quézel, 1999; Brooks et al., 2001; De Klerk et al., 2004; Rodrigues et al., 2004a). IBA coverage of threatened species in these regions was better than PA coverage, although it was still poor relative to IBA coverage in other areas. Expansion of both the PA (and IBA) networks should therefore be considered in these regions.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

This work was carried out during AEB's placement at the RSPB from an MRes at the University of York, organized by Dr Jane Hill. Lucy Arnold of the RSPB provided technical assistance with GIS. We thank L. Roxburgh and R. Safford for unpublished information. We are grateful to the many thousands of experts and organizations, both within and beyond the BirdLife Partnership, who contribute to the development of extensive datasets on threatened species and IBAs, without which the current study would not be possible. Finally, we thank the two anonymous referees whose comments helped to greatly improve the manuscript.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information
  • BirdLife International (2008a). State of the world's birds: indicators for our changing world. Cambridge: BirdLife International.
  • BirdLife International (2008b). Threatened birds of the world 2008. CD-ROM. Cambridge: BirdLife International.
  • Boyd, C., Brooks, T.M., Butchart, S.H.M., Edgar, G.J., da Fonseca, G.A.B., Hawkins, F., Hoffmann, M., Sechrest, W., Stuart, S.N. & van Dijk, P.P. (2008). Scale and the conservation of threatened species. Conserv. Lett. 1, 3743.
  • Brooks, T., Balmford, A., Burgess, N., Hansen, L.A., Moore, J., Rahbek, C., Williams, P., Bennun, L., Byaruhanga, A., Kasoma, P., Njororge, P., Pomeroy, D. & Wondafrash, M. (2001). Conservation priorities for birds and biodiversity: do East African Important Bird Areas represent species diversity in other terrestrial vertebrate groups? Ostrich 15 (Suppl.): 312.
  • Brooks, T.M., Mittermeier, R.A., da Fonseca, G.A.B., Gerlach, J., Hoffmann, M., Lamoreux, J.F., Mittermeier, C.G., Pilgrim, J.D. & Rodrigues, A.S.L. (2006). Global biodiversity conservation priorities. Science 313, 5861.
  • Bruner, A.G., Gullison, R.E., Rice, R.E. & Fonseca, G.A.B. (2001). Effectiveness of parks in protecting tropical biodiversity. Science 291, 125128.
  • Buchanan, G.M., Butchart, S.H.M., Dutson, G., Pilgrim, J.D., Steininger, M.K., Bishop, K.D. & Mayaux, P. (2008). Using remote sensing to inform conservation status assessment: estimates of recent deforestation rates on New Britain and the impacts upon endemic birds. Biol. Conserv. 141, 5666.
  • Butchart, S.H.M., Walpole, M., Collen, B., van Strien, A., Scharlemann, J.P.W., Almond, R.E.E., Baillie, J.E.M., Bomhard, B., Brown, C., Bruno, J., Carpenter, K.E., Carr, G.M., Chanson, J., Chenery, A.M., Csirke, J., Davidson, N.C., Dentener, F., Foster, M., Galli, A., Galloway, J.N., Genovesi, P., Gregory, R.D., Hockings, M., Kapos, V., Lamarque, J-F., Leverington, F., Loh, J., McGeoch, M.A., McRae, L., Minasyan, A., Morcillo, M.H., Oldfield, T.E.E., Pauly, D., Quader, S., Revenga, C., Sauer, J.R., Skolnik, B., Spear, D., Stanwell-Smith, D., Stuart, S.N., Symes, A., Tierney, M., Tyrrell, T.D., Vié, J.C. & Watson, R. (2010). Global biodiversity: indicators of recent declines. Science 328, 11641168.
  • Coad, L., Burgess, N.D., Loucks, C., Fish, L., Scharlemann, J.P.W., Duarte, L. & Besançon, C. (2010). Reply to Jenkins and Joppa – expansion of the global terrestrial protected area system. Biol. Conserv. 143, 56.
  • De Klerk, H.M., Fjeldså, J., Blyth, S. & Burgess, N.D. (2004). Gaps in the protected area network for threatened Afrotropical birds. Biol. Conserv. 117, 529537.
  • Donald, P.F., Aratrakorn, S., Htun, T.W., Eames, J.C., Hla, H., Thunhikorn, S., Sribua-Rod, K., Tinun, P., Aung, S.M., Zaw, S.M. & Buchanan, G.M. (2009). Population, distribution, habitat use and breeding of Gurney's Pitta Pitta gurneyi in Myanmar and Thailand. Bird Conserv. Int. 19, 353366.
  • Dudley, N. (Ed.). (2008). Guidelines for applying protected area management categories. Gland: IUCN.
  • Fishpool, L.D.C. & Evans, M.I. (2001). Important Bird Areas in Africa and associated islands: priority sites for conservation. Newbury/Cambridge: Pisces Publications/BirdLife International.
  • Gaston, K.J. & Fuller, R.A. (2009). The size of species' geographic ranges. J. Appl. Ecol. 46, 19.
  • del Hoyo, J., Elliot, A. & Sargatal, J. (1992). Handbook of the birds of the world. Barcelona: Lynx Edicions.
  • IUCN. (2008). IUCN Red List of threatened species. Gland: IUCN. Available at http://www.iucnredlist.org (accessed 18 August 2008).
  • Jenkins, C.N. & Joppa, L. (2009). Expansion of the global terrestrial protected area system. Biol. Conserv. 142, 21662174.
  • Jenkins, C.N. & Joppa, L. (2010). Considering protected area category in conservation analyses. Biol. Conserv. 143, 78.
  • Jetz, W., Sekercioglu, C.H. & Watson, J.E.M. (2008). Ecological correlates and conservation implications of overestimating species geographic ranges. Conserv. Biol. 22, 110119.
  • Joppa, L.N., Loarie, S.R. & Pimm, S.L. (2008). On the protection of “protected areas”. Proc. Natl. Acad. Sci. USA 105, 66736678.
  • Kear, J. (2005). Ducks, geese and swans. Oxford: Oxford University Press.
  • Mayaux, P., Bartholomé, E., Fritz, S. & Belward, A. (2004). A new land-cover map of Africa for the year 2000. J. Biogeogr. 31, 861877.
  • Médail, F. & Quézel, P. (1999). Biodiversity hotspots in the Mediterranean basin: setting global conservation priorities. Conserv. Biol. 13, 15101513.
  • Mwangi, M.A.K., Butchart, S.H.M., Barasa, F., Bennun, L.A., Evans, M.I., Fishpool, L.D.C., Kanyanya, E., Madindou, I., Machekele, J., Matiku, P., Mulwa, R., Stattersfield, A.J. & Siele, J. (in press) Tracking trends in key sites for biodiversity: a case study using Important Bird Areas in Kenya. Bird Conserv. Int.
  • Nagendra, H. (2008). Do parks work? Impact of protected areas on land cover clearing. Ambio 37, 330337.
  • Osborne, P.E., Alonso, J.C. & Bryant, R.G. (2001). Modeling landscape-scale habitat use using GIS and remote sensing: a case study with great bustards. J. Appl. Ecol. 38, 458471.
  • Pressey, R.L., Humphries, C.J., Margules, C.R., Vane-Wright, R.I. & Williams, P.H. (1993). Beyond opportunism: key principles for systematic reserve selection. Trends Ecol. Evol. 8, 124128.
  • Rodrigues, A.S.L., Akçakaya, H.R., Andelman, S.J., Bakarr, M.I., Boitani, L., Brooks, T.M., Chanson, J.S., Fishpool, L.D.C., da Fonseca, G.A.B., Gaston, K.J., Hoffmann, M., Marquet, P.A., Pilgrim, J.D., Pressey, R.L., Schipper, J., Sechrest, W., Stuart, S.N., Underhill, L.G., Waller, R.W., Watts, M.E.J. & Yan, X. (2004a). Global gap analysis: priority areas for expanding the global protected-area network. BioScience 54, 10921100.
  • Rodrigues, A.S.L., Andelman, S.J., Bakarr, M.I., Boitani, L., Brooks, T.M., Cowling, R.M., Fishpool, L.D.C., da Fonseca, G.A.B., Gaston, K.J., Hoffmann, M., Long, J.S., Marquet, P.A., Pilgrim, J.D., Pressey, R.L., Schipper, J., Sechrest, W., Stuart, S.N., Underhill, L.G., Waller, R.W., Watts, M.E.J. & Yan, X. (2004b). Effectiveness of the global protected area network in representing species diversity. Nature 428, 640643.
  • Rodrigues, A.S.L. & Gaston, K.J. (2001). How large do reserve networks need to be? Ecol. Lett. 4, 602609.
  • Rondinini, C. & Boitani, L. (2006). Differences in the umbrella effects of african amphibians and mammals based on two estimators of the area of occupancy. Conserv. Biol. 20, 170179.
  • Rondinini, C., Chiozza, F. & Boitani, L. (2006a). High human density in the irreplaceable sites for African vertebrates conservation. Biol. Conserv. 133, 358363.
  • Rondinini, C., Stuart, S. & Boitani, L. (2005). Habitat suitability models and the shortfall in conservation planning for African vertebrates. Conserv. Biol. 19, 14881497.
  • Rondinini, C., Wilson, K.A., Boitani, L., Grantham, H. & Possingham, H.P. (2006b). Tradeoffs of different types of species occurrence data for use in systematic conservation planning. Ecol. Lett. 9, 11361145.
  • SCBD. (2010). Global biodiversity outlook 3. Montreal: Secretariat of the Convention on Biological Diversity.
  • Stattersfield, A.J., Crosby, M.J., Long, A.J. & Wege, D.C. (1998). Endemic bird areas of the world. Cambridge: Bird Life International.
  • UN (2010). The millennium development goals report. New York: United Nations.
  • UNEP-WCMC (2007). World Database on Protected Areas. Available at http://www.unep-wcmc.org/wdpa/index.htm (accessed 25 July 2008).
  • USGS (2008). Earth Resources Observation and Science (EROS). Available at http://eros.usgs.gov/products/elevation/gtopo30/gtopo30.html (accessed 2 June 2008).

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Table S1. Conversion between IUCN Habitat Classification Scheme and Land Cover Map of Africa habitat classifications.

Table S2. 2008 IUCN Red List category, Extent of Occurrence (EOO), Extent of Suitable Habitat (ESH), percentage of EOO containing suitable habitat, percentages of ESH within Important Bird Areas (IBAs) and Protected Areas (PAs), and number of IBAs designated for each species for the 157 species of threatened terrestrial African birds covered in this study. VU=Vulnerable, EN=Endangered, CR=Critically Endangered. Conservation efforts for species with EOOs>100 000 km2 (marked *) are likely to require a greater emphasis on broad-scale, rather than site-based, conservation actions.

Table S3. Numbers of category A1 and other Important Bird Areas (IBAs) in mainland Africa and Madagascar included and excluded (due to lack of digitized boundaries) from the study. Category A1 IBAs regularly hold significant numbers of a globally threatened species, or other species of global conservation concern.

Figure S1. Histogram of percentage of Extent of Suitable Habitat (ESH) within Protected Areas (PAs) for 144 globally threatened African bird species.

Figure S2. Density of species based on Extent of Suitable Habitat (ESH) for globally threatened African bird species with <25% of their ESH within Important Bird Areas, mapped as maximum number of species per 50 km2 to increase clarity of figure, and excluding species with Extent of Occurrence >100 000 km2.

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer-reviewed and may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

FilenameFormatSizeDescription
ACV_398_sm_supllinfo.pdf623KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.