SEARCH

SEARCH BY CITATION

Keywords:

  • Administrative units;
  • conservation target;
  • Convention on Biological Diversity;
  • coordinated conservation;
  • spatial conservation prioritization;
  • Zonation software

Abstract

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

Aim

Global conservation policies, such as the Convention on Biological Diversity (CBD) decision to aim for the protection of 17% of the area of terrestrial ecosystems by 2020, are typically realized at national levels. We investigate the difference between continentally coordinated conservation versus nationally devolved conservation, in a manner relevant for the Nagoya resolution.

Location

The terrestrial areas of the Western Hemisphere.

Methods

We used IUCN distribution data for 8463 species of mammals, birds and amphibians in the Western Hemisphere. We investigated the consequences of prioritizing land at a continental scale, versus analysing priorities within each country separately. Spatial prioritization was performed using the Zonation software, which produces a complementarity-based hierarchical priority ranking across the area of interest.

Results

We found that coordinated continent-wide priorities achieved > 50% higher mean protection levels than national analyses for the top 17% of land. National prioritizations also result in spatial priority patterns that can be considered as artefacts at the continental scale: in bands of high-priority land concentrated at terrestrial political boundaries, such as at low-latitude edges of temperate zone countries. We find that this edge artefact also correlates with the present distribution of conservation areas, with the density of conservation areas within 50 km of a national border being > 50% higher than the density of conservation areas away from national borders.

Main conclusions

The means by which national priorities are integrated with continental or global conservation prioritization will have considerable influence on how much is achieved by the CBD resolution. Focus on national species distributions and priorities will result in lost performance because of emphasis on nationally rare species that are comparatively common elsewhere. National borders intersect species distributions (and possibly diversity gradients), leading to clustering of nationally rare species and priority areas close to the border.


Introduction

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

International conservation strategies and priorities are frequently devolved to national governments and conservation organizations for their implementation. This facilitates delivery within national legal systems and allows international priorities to be adjusted to the needs of local biological and human communities. Conservation priorities are currently particularly relevant with respect to the implementation of the resolutions of the tenth Conference of Parties to the Convention on Biological Diversity, which was held on October 2010 in Nagoya, Japan. It was decided that biodiversity loss should be halted and that biodiversity should be valued, conserved, restored, wisely used and shared equitably, maintaining ecosystem services, sustaining a healthy planet and delivering benefits essential for all people (Normile, 2010; UNEP/CBD, 2010; Harrop, 2011). Global efforts towards a sustainable environment will be investigated further in the Rio + 20 conference (UNCSD, 2012).

It has long been argued that the prior 10% global conservation target was inadequate, in part because of inefficient site selection whereby existing protected areas are often located in low-productivity and inaccessible regions, leading to imbalances in the representation of different ecosystems in the global protected area network (Rodrigues et al., 2004; Gaston et al., 2008; Fuller et al., 2010). One of the most prominent vehicles of action arising from the Nagoya meeting was Target 11, a decision to increase the area of terrestrial ecosystems protected to 17% globally (Harrop, 2011). This target has already been partway met in terms of area, as the present global percentage of protected area has increased to over 12% (McDonald & Boucher, 2011). Nevertheless, given prior massive imbalances in protection levels across biomes, there necessarily remains much need for expanded protection measures (Jenkins & Joppa, 2009). The Nagoya meeting also called for conservation ‘decision-making [to be] based on sound science’ (UNEP/CBD, 2010) – in the context of the 17% target, how can priority areas for conservation be selected efficiently to protect the maximum amount of biodiversity possible in a given area?

The requirement of equitability suggests that conservation effort should take place in all countries (UNEP/CBD, 2010), but previous work suggests that globally planned conservation would be more efficient than locally distributed conservation effort. Soutullo & Gudynas (2006) investigated protection of bioregions across the countries of South America and found that regionally coordinated conservation would be more efficient than national decision-making. Much of the Earth's unprotected biodiversity lies in global priority areas identified by conservation NGOs, which also suggests a need for globally coordinated action (Soutullo et al., 2008). In the USA, a spatially explicit study found that representing mammal species in all states separately required ten times the area than representing species nationally (Vazquez et al., 2008). Coordinated conservation of amphibians, reptiles and fish around the Mediterranean would have been 45% more efficient than uncoordinated conservation (Kark et al., 2009). Coordinated conservation at European wetlands would have been 30% more efficient than uncoordinated effort (Jantke & Schneider, 2010).

Going beyond international and national, Pajaro et al. (2010) find that policy development related to marine protected areas occurs at three levels: international, national and local, with information feedback and conflict resolution between levels needed. It has also been suggested that marine conservation planning should be integrated with broader marine spatial planning and ocean zoning to avoid failure of conservation effort (Agardy et al., 2011). Inside a country, Strange et al. (2006) investigated conservation within Denmark and found that nationally coordinated conservation could be up to 20 times more efficient than regionally devolved effort. They suggest that in economic and biodiversity terms, it can largely be a win–win situation to set a common goal, to develop priority strategies and to coordinate actions at higher rather than at lower levels of administration. Chiarucci et al. (2008) investigated patterns of species composition across the European Natura 2000 network and found that ecosystem-level complementarity of areas is critical for the performance of the network – and achieving such complementarity at the European scale requires coordinated conservation effort. Indeed, in Europe, conservation has been moving towards a higher level of international coordination (Cogalniceanu & Cogalniceanu, 2010). Taking a more economical perspective, White et al. (2012) found that management of ecosystem services that is coordinated across interacting sectors and stakeholders may produce more than double the societal gains that what can be expected from uncoordinated effort (White et al., 2012). In the Netherlands, collective contracts allow neighbouring land managers to coordinate environmental management at the landscape rather than the farm-scale, reducing costs and increasing participation rates in conservation (Franks, 2011).

Following the Nagoya call for a science-based approach, we here present new spatially explicit large-scale analyses that contrast national and continental targeting of conservation, specifically in the context of the 17% terrestrial conservation coverage target. We use ‘continental’ as a shorthand for analyses where priorities are evaluated in a coordinated manner across one or more continents – here, North, Central and South America, and associated islands were used as a single entity for analysis. We use ‘national’ as a shorthand for analyses across the same overall geographical region, but where independent priorities are devolved, and hence set independently for each country within these continents. We in particular focus on what can be achieved with 17% of land when it is selected in a continentally efficient manner, compared to selection that emphasizes approximately equal conservation efforts across all countries. We also focus on what happens around the borders of countries, as previous work has shown that the boundary of the planning region influences both the total area needed to meet conservation goals and the spatial location of suggested additions to conservation area networks (Huber et al., 2010). The same general principles and issues identified here apply to levels of selection other than 17%.

Methods

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

Data

We undertook a continental-scale priority ranking analysis for the terrestrial areas of the Western Hemisphere. Analysis was based on the distributions of 2928 amphibians from the world amphibian data (IUCN, Conservation International, & NatureServe, 2006), 1642 mammals from the world mammal data (IUCN, 2009), and information on 3893 bird species obtained from Ridgely et al. (2007). The distribution polygons of the amphibian and mammal species were converted into equal-area presence–absence occupancy grids (World Behrmann projection, datum WGS 1984; 50 km × 50 km rectangular grid cells), using the ‘Vector to Grid’ function in ARCGIS/ARCINFO 9.2 (ESRI, Redlands, CA, USA). Marine species and grid cells were excluded from this analysis. Islands that were too small to include the centre point of even one grid cell were also excluded. We calculated existing levels of protection for each grid cell by calculating the fraction of the grid cell covered by land designated as IUCN categories I-IV protection levels, using the IUCN & UNEP, 2010 data (IUCN & UNEP, 2010).

Analysis methods

We used a publicly available spatial conservation prioritization method and software, Zonation, to produce large-scale conservation priority rankings (Moilanen et al., 2005; Arponen et al., 2012). The principles of these analyses used here have been described in detail elsewhere, and the software and documentation are freely publicly available (Moilanen et al., 2005, 2011a,b). In the analyses presented here, we evaluate the efficiency with which (and locations where) species could be protected, although we could also consider other entities of importance, such as ecosystem services (Moilanen et al., 2011a).

Verbally described, spatial priority ranking by Zonation takes a series of units, here equal-area grid cells, and iteratively ranks them in order, from lowest to highest priority, based on the representation of biodiversity and other information from those cells. Normally, this method is applied to all of the cells in one region (e.g. country), for example, to identify the most important locations for conservation. Here, we use an analysis variant that effectively joins multiple conservation prioritization analyses, one global and one for each administrative sub-region (Moilanen & Arponen, 2011). This analysis is best understood by examining an intermediate equation that defines how conservation value is handled during the ranking process (Moilanen & Arponen, 2011):

  • display math(1)

This equation gives conservation value V() as a function of a set of areas, S. Value is combined from two components, a global (here continental) one and a local (national) one, the balance between which is tuned by parameter q. In this equation, weights (w), benefit functions (f) or representation (R) are indexed by administrative sub-region (index A), by species (j), and for global (G) and local (L) components. During iterative ranking, change in conservation value V(S) is evaluated for loss of each grid cell remaining in S, and the cell that leads to smallest loss is chosen for removal next. Starting from the full landscape and iteratively minimizing loss leads to a ranking that identifies least important grid cells at low ranks and successively more valuable cells are retained for higher ranks.

Features relevant for the present analysis include: (1) setting q = 0 produces an analysis where rankings for individual countries are effectively independent from the rankings of other countries – this is our national analysis. (2) Setting = 1 results in an analysis where only global considerations influence the ranking – this is our continental analysis. (3) An intermediate q between 0 and 1 produces a compromise analysis that combines global and local considerations. After generating a ranking, we can identify its top 17% (or any other percentage) of cells, corresponding to the Nagoya Target. The top 17% can also be identified separately from within each country, equivalent to national devolution of Target 11. With respect to the compromise solution, the one we used effectively allows deviation from the per-country 17% to allow higher emphasis on species-rich tropical regions. A different kind of compromise would be to select 17% for each country, but to do so based on continental priorities. This result could be obtained by picking the highest-ranked 17% of cells for each country from the continental analyses.

We repeated each of the above analyses (continental, national, compromise) using two major Zonation analysis variants, the additive benefit function (ABF) and core-area Zonation (CAZ) (Moilanen et al., 2005, 2011a). ABF and CAZ represent conceptually different views of conservation value, and hence, the analyses complement each other. ABF favours grid cells that contain large numbers of localized species (summing value across species), combined with a species–area approach to minimize extinction rates (Moilanen et al., 2011a). The ABF power function was set to z= 0.25 for all species, typical of species–area curves. CAZ considers each species separately, securing high-quality locations for all species, even when they occur in otherwise species-poor regions (Moilanen et al., 2005, 2011a). Technically, differences between ABF and CAZ manifest in different across-species aggregation structures and different benefit functions in equation 1; see e.g. Moilanen et al. (2011a).

With respect to parameter values, we treated all species as equal, with species weights w= 1.0 for all species j, both locally and globally. [Note that even when equally weighted, Zonation emphasizes narrow-range and shrinking range species because of iterative range-size normalization employed in computations, maintaining a balance across all features all through the ranking (Moilanen et al., 2005, 2011a).] The weights of countries were set equal to the area of the country (Table 1) and normalized to sum to 1 (Moilanen & Arponen, 2011). We used = 0.003 (ABF) and = 0.001 (CAZ) in the compromise analyses, as values that achieved approximately intermediate performance between continental coordinated and national analyses (= 0.5 does not imply an analysis that is exactly balanced between global and local units as the interpretation of q is case-specific and depends, among other things, on the weights given to countries – see Moilanen & Arponen, 2011).

Table 1. Statistics about countries included in this study, arranged according to the global CAZ analysis (by percentage of country assigned to top 17%). Statistics are not shown for the national solutions because 17% of each country will be included in its top 17%. The column ‘CBD’ shows whether the country is a member of the Convention on Biological Diversity, with consequent implications on the implementation of the Nagoya resolutions in the country
CountryNo. of cells% of cells of the country in top 17% fractionCBD member
ABFCAZ
ContinentalCompromiseContinentalCompromise
Haiti4100.050.0100.050.0No
Jamaica2100.0100.0100.0100.0No
Costa Rica17100.082.4100.047.1Yes
Puerto Rico2100.0100.0100.0100.0No
Panama20100.080.095.040.0Yes
Ecuador87100.072.486.231.0Yes
Cuba29100.037.979.337.9Yes
Dominican Rep.17100.052.976.535.3Yes
Guatemala4372.137.269.834.9Yes
Honduras4386.032.667.423.3Yes
Belize8100.012.562.50.0Yes
Peru49567.137.644.225.3Yes
Chile22330.59.938.112.1No
Nicaragua4288.119.038.17.1Yes
Colombia43842.536.837.724.9Yes
French Guiana3287.515.637.56.3No
Mexico72738.424.237.427.1Yes
Venezuela35541.124.233.520.3Yes
Uruguay7219.45.629.24.2Yes
El Salvador7100.028.628.614.3Yes
Suriname5944.110.225.46.8Yes
Bolivia43526.922.825.314.5Yes
Argentina110113.213.420.417.3No
Guyana8226.814.618.34.9Yes
Brazil334816.626.415.222.5Yes
Paraguay16010.68.813.11.3Yes
United States35788.011.311.917.9No
Canada34630.44.72.47.1No

The present work bears superficial resemblance to those employing the ‘maximum coverage principle’ in reserve selection (Camm et al., 2002; Margules & Sarkar, 2007). These studies employ optimization to maximize the number of (species-level) representation targets that can be satisfied given limited resources and concentrate only on the best parts of the study area. Zonation represents very different principles to those employed in target-based systematic conservation planning. The continuous ranking of locations (lowest to highest value) that we produce does not require us to set a priori area or species targets or to concentrate only on the ‘best cells’ (Moilanen et al., 2005). It is practically and quantitatively advantageous that targets need not be set separately for 8463 features (species) in each of the 32 countries (Di Minin & Moilanen, 2012). Nonetheless, we can still derive the locations of the ‘best’ 17% once the analysis is complete, facilitating interpretation in the context of the Nagoya resolutions.

Results

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

Spatial prioritization at the continental-scale results in most high-priority areas being located at low latitudes and often in montane and insular regions (Fig. 1a,b). When analyses are carried out in each country separately, priority areas are evenly distributed among countries, the red and orange colours identifying the top 17% within each country separately (Fig. 1c,d). Most of Central America and countries along the Andes receive high conservation priorities in continental-scale analyses (Table 1) but, by definition, only 17% of their land would be prioritized within national-level analyses. Prioritizations that simultaneously implement joint continental and national priorities generate intermediate patterns (Fig. 2, Table 1). ABF and CAZ analyses lead to similar conclusions, albeit with a ‘spottier’ distribution within the CAZ solutions (Figs 1 & 2), implying that while most species could be efficiently protected by large semi-continuous conservation areas, there are species that only occur in relatively species-poor regions critical for few species only.

image

Figure 1. Conservation prioritization for the Western Hemisphere based on amphibians, mammals and birds. We show results for continental ABF and CAZ analyses (a and b), and national ABF and CAZ analyses (c and d). The continental prioritizations have treated the entire Western Hemisphere as one planning unit, with the oranges and reds identifying the highest priority 17% of land across the entire region. The devolved, national analyses are based on the national distributions of species, with oranges and reds identifying the highest priority 17% of land within each country, ignoring the distributions of species elsewhere in North and South America. ABF places higher emphasis on species richness than CAZ, which aims at the inclusion of high-quality areas for all species even when they occur in otherwise species-poor regions.

Download figure to PowerPoint

image

Figure 2. Combined continental + national compromise prioritizations for ABF (a) and CAZ (b). By varying the relative weight given to the continental component, it is possible to generate an effectively continuous range of compromise solutions between the continental and national solutions (Fig. 1).

Download figure to PowerPoint

National-scale analyses result in major artefacts because of the imposition of political boundaries. This is particularly striking along the border between Canada and the contiguous USA, where adjacent cells that contain virtually identical biotas are top 17% cells on the Canadian side of the border but in some cases bottom 17% cells in the USA (Fig. 1). The thermal-latitudinal diversity gradient results, in this case, in similar biotas having high priority in Canada but low priority in the USA. Fundamentally, the cause of this artefact is that species that are rare in one country (say, Canada) are abundant in another (say, USA). Similar effects are seen at other temperate zone borders, particularly the USA–Mexico border, and to some extent in northern Argentina, where the national-scale analyses cluster priority areas near the countries' low-latitude borders. These artefacts extend to the tropics, with national-level analyses, for example, showing increased representation close to the geographical borders of Brazil (e.g. in the north with French Guiana and in the west with Peru). In each case, areas are identified that are nationally more important than they would be at a continental scale.

Zonation performance curves quantify the fractions of species' distributions (here averaged across species) that are covered at any stage of the priority ranking. These show that continental-scale analyses are most efficient, and national-only analyses are least efficient, at delivering species-level global biodiversity targets (Fig. 3). ABF is on average more efficient than CAZ (Fig. 3), but the latter increases the cover given to localized species that occur in low-diversity regions (a result that may be desired but is not visible in averaged performance curves). Considering the top 17% of the land alone, ABF delivers 66%, and CAZ 60.5%, of the distribution areas (averaged across species) of Western Hemisphere mammals, birds and amphibians in the continental-scale analyses (Fig. 3). Because a random selection of 17% of area would by statistical necessity provide 17% mean coverage across species, it follows that efficient continental selections are 3.9 and 3.6 times as effective as selecting land at random, for ABF and CAZ, respectively. The continental ABF analysis provides an upper limit to the biodiversity that could potentially be obtained in 17% of the terrestrial area (Figs 1 & 3). In contrast, the national ABF or CAZ solutions provide only 42% and 39% mean coverage across species distributions. While ABF compromise performance falls smoothly between continental and national analyses, the CAZ compromise solution performs poorly, particularly in identifying the top 10% of the land surface (Fig. 3). This somewhat surprising result is because the CAZ compromise must retain high-quality locations for all species both locally and globally, which requirement is more stringent than providing only continental or national coverage. In effect, covering all nationally or continentally rare species leads to losses in mean conservation levels across all species as a whole.

image

Figure 3. The Zonation performance curves corresponding to the priority maps of Figures 1 and 2. These curves report the mean (across species) fraction of the distribution of each species retained as a function of fraction of land reserved for protection. Of these curves, the continental ABF solution is most efficient: it is able to retain the highest fraction of species distributions, implying high return on investment in areas belonging to high ABF ranking cells. CAZ produces slightly lower mean performance, compared to ABF, because CAZ gives higher protection levels to species occurring in relatively species-poor areas. All variants of local (national) prioritization lose efficiency. This is because conservation priorities are relocated from endemic- and species-rich tropical and montane regions to relatively species-poor areas closer to the poles. Differences in conservation efficiency between methods are substantial: for example, national-level CAZ analysis provides < 40% mean coverage across all species within the top 17% of the land, whereas continental ABF provides ~66% coverage.

Download figure to PowerPoint

We tested the stability of our analyses to changes in species data by two approaches. First, we considered the robustness of conclusions to choice of taxonomic group (see Appendix S1 in Supporting Information), finding that our main conclusions do not change. For example, very strong bands of high-priority land are present in national analyses and absent in continental analyses along all of the borders where this artefact was identified by the main analyses (Fig. 1; Fig. S1, Fig. S2, Fig. S3). In contrast to previous work (Kark et al., 2009), the edge artefact is also present for amphibians for both temperate zone and tropical borders. These separate analyses also identify that some regions are of high priority for all three taxonomic groups. Such areas include the Atlantic rain forest of Brazil, southern parts of Mexico, the spine of the Andes and parts of the west coast of USA. Some regions are of higher significance for individual taxa, but most of the areas that are priorities for one of the taxonomic groups are also prioritized for at least one of the other groups (Table S1; e.g. Hispaniola is relatively more important for birds and amphibians than for mammals, and the Amazon River corridor emerges more strongly for birds and mammals than for amphibians). Continental analyses consistently place higher priority on large areas of Central America, on the west coast of USA and on areas of South America expanding either side of and down the spine of the Andes, whereas national analyses tend to redistribute priorities towards national boundaries.

We also tested the stability of our analyses to the number of species included using sensitivity analysis. The basic continental and national ABF and CAZ analyses (Fig. 1) were replicated 10 times taking a random 50% sample of the full 8463 species. It was found that the top 17% fractions of the landscape on average overlapped to a degree of 84.8% (national ABF; SD = 0.45%), 85.3% (continental ABF; SD = 0.37%), 69.8% (national CAZ; SD = 1.6%) and 72.8% (continental CAZ; SD = 1.2%). Thus, coordinated continental-scale analyses were robust to the particular species included.

We also examined the relationship between the ranking of locations in the Zonation analyses and the distribution of actual protected areas (PAs), defined as existing IUCN category I-IV PAs. PAs are not very efficiently placed by these criteria: the very highest priority grid cells according to Zonation ranks do have slightly higher fractions protection than land overall, but conservation areas are also widely located across lowest ranked areas (Fig. 4). Only approximately 10% of the top 5% of land is protected, indicating that any unprotected natural or semi-natural habitat within the top-priority cells could be considered for protection as the world works towards fulfilling the Nagoya 17% target. The virtual lack of correlation between continental-scale ranked priorities and existing PAs (Fig. 4) indicates, and agrees with prior evaluations (Gaston et al., 2008), that conservation designation is inefficient (from a biodiversity perspective), although of course PAs are likely be situated in better-than-average habitat inside each cell.

image

Figure 4. Average fraction of grid cells protected compared to priority rank in the present Zonation analyses (with 1% intervals on x-axis). Rank 99–100 corresponds to the top 1% and rank 0–1 to the least important 1%.

Download figure to PowerPoint

The role of national priorities in past PA selection can be seen in two ways. First, there are stronger (albeit still quite weak) correlations between national priority ranks and the designation of PAs, than for continental priorities (Fig. 4). Furthermore, the relationship is stronger for (the overall less efficient) CAZ, suggesting an emphasis on prioritizing individual species. Second, existing PAs disproportionately cluster around country boundaries. Comparison of the top 17% areas (as defined by the ABF or CAZ analyses) against the distribution of present protected areas identifies areas where conservation would be highly efficient at the continental level (yellow–orange colour, Fig. 5a,b). The artefact that national priorities cluster around terrestrial borders between countries (Fig. 5c,d) is seen by the concentration of top 17% priority cells around country borders (Fig. 6). Bias towards borders is actually also seen in the present distribution of IUCN category I-IV protected areas: 11.9% of the continents' PA land currently falls within border cells, which comprised only 7.7% of the continental land surface in our analysis.

image

Figure 5. Potential for expanded conservation measures. Colours from yellow to red show the fraction of land currently protected in the top 17% areas of priority rankings. The greyscale shows the fraction of land already protected in areas that do not belong to top 17% regions. Colours in light yellow and orange show areas that belong to top 17% ranks but where present conservation coverage is below 2%. Results are shown both for the ABF and CAZ, continental and national analyses corresponding to Fig. 1.

Download figure to PowerPoint

image

Figure 6. Distributions of Zonation priority ranks at grid cells adjacent to national terrestrial borders between countries. Points are in one per cent bins, by Zonation rank. In the continental analyses, the highest 17% conservation-value (rank 83–100) cells are no more likely than mid-ranking cells to occur close to political borders. In contrast, high ranks are disproportionately associated with borders in national analyses. The very lowest ranks do not occur at borders in the continental analyses because they are located in comparatively species-poor areas of northern Canada where there are no national borders.

Download figure to PowerPoint

Discussion

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

The CBD Nagoya Target 11 has again raised the profile of where best to protect ecosystems, leading to choices of where protected areas (PAs) and other conservation-related land designations and actions (e.g. conservation easements) should best be located. One, but only one, of the considerations is where biodiversity can be conserved most effectively, as a cultural ecosystem service in its own right, and for its contributions to other ecosystem goods and services and their resilience. The analyses presented here contribute to this aspect of decision-making and need to be judged alongside a number of other social, economic and political interests (Ceballos & Ehrlich, 2006; Wilson et al., 2007; Nelson et al., 2009; Eklund et al., 2011). Given that decisions that flow from CBD targets, including legislative designation of PAs, are commonly devolved to individual countries, each country must judge how best to balance national versus international interests in the implementation of targets. The results in this article highlight that this choice, as applied to the mammal, bird and amphibian biota of the New World, has major implications as to where priority areas lie, as well as for the overall efficiency of conservation strategies. While the targets discussed in Nagoya were determined in a collaborative process, each country may still act independently to achieve the target within its national boundaries, which will likely lead to redundancy and inefficiency in conservation efforts. Prior evidence from conservation biogeography suggests that the need for additional conservation differs greatly across biomes and countries, also suggesting that equal national conservation area targets may be ecologically inefficient (e.g. Jenkins & Joppa, 2009; Schuldt & Assmann, 2010; Marinesque et al., 2012). In this study, the coordinated, continental-scale analyses delivered over one and a half times the conservation value (measured as the average fraction of species' ranges protected) of the nationally devolved analyses.

National-scale decision-making can potentially assign high priorities to locations that are unusual in that country but of low international importance. The analyses we present show that national-only prioritization is inefficient at protecting biodiversity internationally, but that this can be improved somewhat in ‘compromise’ solutions that take both national and international interests into account. However, the results point not just to overall inefficiencies of fully devolved decision-making, but also to major artefacts. Country-level prioritization may result in the clustering of highly ranked areas near to political boundaries. This observation was previously made by Vazquez et al. (2008) for mammals of North America, but is strikingly visible across the larger area analysed in the present work. There are two other studies that have investigated national versus coordinated conservation. Jantke & Schneider (2010) did not observe high priorities close to geographical boundaries, probably because the environment analysed, wetlands of Europe, has a very fragmented distribution. Neither were national edge artefacts observed in another study that used data about distributions of amphibians, reptiles, freshwater fish and land cover types around the Mediterranean basin (Kark et al., 2009). Our results suggest that these artefacts are strong, that they apply to tropical national boundaries as well as to those in the temperate zone (Fig. 1), and that they seriously reduce the overall efficiency of conservation strategies.

The clustering of national, but not international, priorities near political boundaries arises because arbitrarily located country boundaries (from the perspective of biodiversity) cut across the geographical ranges of species. Hence, a single species may be localized near to the geographical edge of one country (where national conservation agencies may prioritize it), but potentially widespread in another country. If so desired, boundary effects might be partially compensated in national analysis by prioritizing the conservation of internationally rare or endangered species within countries. Nevertheless, the boundary artefact will still arise if nationally, but not internationally, rare species are also prioritized within countries (Fig. 2). Although our analyses involved the distributions of species, similar considerations apply to designations based on nationally (but not internationally) rare habitats, ecosystems, and in some cases, ecosystem goods and services. On the other hand, the higher density of present IUCN protected areas near boundaries of countries might be influenced by additional factors. Observations from economic and institutional geography suggest that patterns of comparatively low habitat degradation may be found along the no-man's land of border zones, making nature conservation in those areas attractive (Dudley et al., 2002). Nonetheless, it is important to recognize that the boundary effects reported here result in an overall reduction in conservation efficiency, at a continental scale.

We tested the robustness of our results in by considering the taxonomic groups (mammals, birds, amphibians) separately and found that the same general issues were relevant. The taxa largely shared the similar priority areas, and the boundary effects were as prominent within each group as it had been for the combined analysis (online Appendix). We also tested the stability of our analyses by replicating them using 50% subsamples of the original data and found that conclusions remained unchanged. It is possible though, that addition of new higher taxa, such as plants or insects, could change priority patterns more significantly, particularly if they differ in environmental constraints and/or average geographical range sizes. The expectation was that ABF analyses should be comparatively stable, as they are more focused on species richness and allow for some compensation between species (Moilanen, 2007). This expectation was borne out by the sensitivity analysis. Overall, ABF top areas (Fig. 1a) show a remarkable degree of connectedness at this coarse resolution, suggesting that the highlighted areas would be critical for continental-scale green ‘infrastructure’ (Tzoulas et al., 2007). In contrast, the CAZ analyses are somewhat more sensitive to the identities of individual species, as they have an aim of securing high-quality locations for all species. This makes CAZ analyses less stable at the national scale (see 'Discussion') because the discovery/addition of new species can generate new ‘nationally rare’ species, that shift the priority areas towards those (usually border) areas. The same conclusion extrapolates to traditional systematic conservation planning, which is focused on implementation of conservation and cost-efficient covering of species-specific representation targets (e.g. Margules & Sarkar, 2007; Pressey & Bottrill, 2008; Sarkar & Illoldi-Rangel, 2010).

It is important to ask what data conservation prioritization should be based on (e.g. Margules & Sarkar, 2007; Wilson et al., 2007; Boitani et al., 2011). It would be desirable to include additional taxa, such as plants and insects, in future analyses but the overall concentration of priority areas in Middle America, the Andes, and the Atlantic forest of Brazil are important for a wide range of taxa (Grenyer et al., 2006; Lamoreux et al., 2006). Given the nature of the biodiversity data, we restricted our analyses to 50-km resolution, but our general conclusions (relative efficiencies of country- and international-level analyses and displacement of priority areas towards political boundaries) are robust to resolution. However, the 50-km resolution is not sufficient to select sites for on-the-ground targeting of conservation action (Hurlbert & White, 2005; Orme et al., 2006; Hurlbert & Jetz, 2007; Jetz et al., 2007; Hermoso & Kennard, 2012). Rather, highly ranking 50-km grid cells represent regions within which to search for specific locations to protect. A series of additional considerations will become important at this stage, including identifying undisturbed ecosystems and habitat types that contain the highest value species. Within these regions, socio-political concerns and human-caused factors will influence operational conservation decisions (Ceballos & Ehrlich, 2006; Knight et al., 2006; Eklund et al., 2011; Rondinini et al., 2011; Wilson et al., 2011). These include implementation costs, opportunity costs and needs of alternative land uses (Pressey et al., 2007; Wilson et al., 2007, 2011; Moilanen et al., 2011a; Rondinini et al., 2011), threats and vulnerability (Brooks et al., 2006; Wilson et al., 2006, 2011; Pressey et al., 2007; Rondinini et al., 2011) and ecosystem services (Naidoo et al., 2006; Nelson et al., 2009; UNEP/CBD, 2010), which are prominent in the Nagoya resolution. Notwithstanding these additional considerations, we note that only a small amount (~10%) of the top 17% of the current land surface is currently protected by IUCN category I-IV protected areas (Figs 4 & 5), so there is undoubtedly scope to select at least some further areas for conservation within these 50-km regions.

The Nagoya convention relies substantially on voluntary conservation action, in which decisions reflect the outcome of multiple competing interests and pressures (Harrop & Pritchard, 2011). Most of these decisions will take place at a regional, national or sub-national level. The combined continental and local analyses presented here have the potential to increase the efficiency of decision-making by enabling country-level decision-making to take place in the context of international priorities and preferably for nations whose land is of relatively low conservation priority to help support conservation delivery in countries where the need is greatest. In terms of conservation implications, the present work strongly suggests that large-scale coordination of conservation efforts in response of the Nagoya treaty would be highly desirable, instead of prioritization based on national distributions of biodiversity features. Focus on national-level protection of species will lead to significantly reduced conservation effectiveness at the global scale. Our results also confirm that national species-based conservation priorities would tend to concentrate close to terrestrial boundaries between countries. This outcome can be expected even without any special ecological factors elevating priorities around borders. Rather, such edge effects are an unavoidable consequence of national boundaries crossing large-scale patterns of species distributions, leaving some species rare just at one side of the border.

Acknowledgements

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

A.M. and F. M.-P. thank the ERC-StG project GEDA (grant 260393) for financial support. B.J.A. and C.D.T. thank NERC for financial support. A.A. thanks the Academy of Finland grant #250126 and EU FP7 project SCALES #226852 for support. We thank Aija Kukkala and Johanna Kuusterä for technical assistance.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Biosketch
  10. Supporting Information
  • Agardy, T., di Sciara, G.N. & Christie, P. (2011) Mind the gap Addressing the shortcomings of marine protected areas through large scale marine spatial planning. Marine Policy, 35, 226232.
  • Arponen, A., Lehtomäki, J., Leppänen, J., Tomppo, E. & Moilanen, A. (2012) Effects of connectivity and spatial resolution of analyses on conservation prioritization across large extents. Conservation Biology, 26, 294304.
  • Boitani, L., Maiorano, L., Baisero, D., Falcucci, A., Visconti, P. & Rondinini, C. (2011) What spatial data do we need to develop global mammal conservation strategies? Philosophical Transactions of the Royal Society of London B, 366, SI 26232632.
  • 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.
  • Camm, J.D., Norman, S.K., Polasky, S. & Solow, A.R. (2002) Nature reserve site selection to maximize expected species covered. Operations Research, 50, 946955.
  • Ceballos, G. & Ehrlich, P.R. (2006) Global mammal distributions, biodiversity hotspots, and conservation. Proceedings of the National Academy of Sciences USA, 103, 1937419379.
  • Chiarucci, A., Bacaro, G. & Rocchini, D. (2008) Quantifying plant species diversity in a Natura 2000 network: Old ideas and new proposals. Biological Conservation, 141, 26082618.
  • Cogalniceanu, D. & Cogalniceanu, G.C. (2010) An enlarged European Union challenges priority settings in conservation. Biodiversity and Conservation, 19, 14711483.
  • Di Minin, E. & Moilanen, A. (2012) Empirical evidence for reduced protection levels across biodiversity features from target-based conservation planning. Biological Conservation, 153, 187191.
  • Dudley, J.P., Ginsberg, J.R., Plumptre, A.J., Hart, J.A. & Campos, L.C. (2002) Effects of war and civil strife on wildlife and wildlife habitats. Conservation Biology, 16, 319329.
  • Eklund, J., Arponen, A., Visconti, P. & Cabeza, M. (2011) Governance factors in the identification of global conservation priorities for mammals. Philosophical Transactions of the Royal Society B, 366, SI 26612669.
  • Franks, J.R. (2011) The collective provision of environmental goods: a discussion of contractual issues. Journal of Environmental Planning and Management, 54, 637660.
  • Fuller, R.A., McDonald-Madden, E., Wilson, K.A., Carwardine, J., Grantham, H.S., Watson, J.E.M., Klein, C.J., Green, D.C. & Possingham, H.P. (2010) Replacing underperforming protected areas achieves better conservation outcomes. Nature, 466, 365367.
  • Gaston, K.J., Jackson, S.E., Cantu-Salazar, L. & Cruz-Pinon, G. (2008) The ecological performance of protected areas. Annual Review of Ecology, Evolution, and Systematics, 39, 93113.
  • Grenyer, R., Orme, C.D.L., Jackson, S.F., Thomas, G.H., Davies, R.G., Davies, T.J., Jones, K.E., Olson, V.A., Ridgely, R.S., Rasmussen, P.C., Ding, T.S., Bennett, P.M., Blackburn, T.M., Gaston, K.J., Gittleman, J.L. & Owens, I.P.F. (2006) Global distribution and conservation of rare and threatened vertebrates. Nature, 444, 9396.
  • Harrop, S.R. (2011) Living in harmony with nature? Outcomes of the 2010 Nagoya conference of the convention on biological diversity Journal of Environmental Law, 23, 117128.
  • Harrop, S.R. & Pritchard, D.J. (2011) A hard instrument goes soft: the implications of the convention on biological diversity's current trajectory. Global Environmental Change, 21, SI474SI480.
  • Hermoso, V. & Kennard, M.J. (2012) Uncertainty in coarse conservation assessments hinders the efficient achievement of conservation goals. Biological Conservation, 147, 5259.
  • Huber, P.R., Greco, S.E. & Thorne, J.H. (2010) Boundaries make a difference: the effects of spatial and temporal parameters on conservation planning. Professional Geographer, 62, 409425.
  • Hurlbert, A.H. & Jetz, W. (2007) Species richness, hotspots, and the scale dependence of range maps in ecology and conservation. Proceedings of the National Academy of Sciences USA, 104, 1338413389.
  • Hurlbert, A.H. & White, E.P. (2005) Disparity between range map- and survey-based analyses of species richness: patterns, processes and implications. Ecology Letters, 8, 319327.
  • IUCN. (2009) IUCN red list of threatened species. Version 2009.1, Available at: http://www.iucnredlist.org (accessed 5 May 2010).
  • IUCN, Conservation International, and NatureServe (2006) Global amphibian assessment. Available at www.globalamphibians.org, version 1.1. (accessed 8 May 2008).
  • IUCN & UNEP. (2010) The World Database on Protected Areas (WDPA). UNEP-WCMC, Cambridge, UK. www.protectedplanet.net. (accessed 13 March 2011).
  • Jantke, K. & Schneider, U.A. (2010) Multiple-species conservation planning for European wetlands with different degrees of coordination. Biological Conservation, 143, 18121821.
  • Jenkins, C.N. & Joppa, L. (2009) Expansion of the global terrestrial protected area system. Biological Conservation, 142, 21662174.
  • Jetz, W., Sekercioglu, C.H. & Watson, J.E.M. (2007) Ecological correlates and conservation implications of overestimating species geographic ranges. Conservation Biology, 22, 140150.
  • Kark, S., Levin, N., Grantham, H.S. & Possingham, H.P. (2009) Between-country collaboration and consideration of costs increase conservation planning efficiency in the Mediterranean Basin. Proceedings of the National Academy of Sciences USA, 106, 1536815373.
  • Knight, A.T., Cowling, R.M. & Campbell, B.M. (2006) An operational model for implementing conservation action. Conservation Biology, 20, 408419.
  • Lamoreux, J.F., Morrison, J.C., Ricketts, T.H., Olson, D.M., Dinerstein, E., McKnight, M.W. & Shugart, H.H. (2006) Global tests of biodiversity concordance and the importance of endemism. Nature, 440, 212214.
  • Margules, C.R. & Sarkar, S.. (2007) Systematic conservation planning. Cambridge University Press, Cambridge, UK.
  • Marinesque, S., Kaplan, D.M. & Rodwell, L.D. (2012) Global implementation of marine protected areas: Is the developing world being left behind? Marine Policy, 36, 727737.
  • McDonald, R.I. & Boucher, T.M. (2011) Global development and the future of the protected area strategy. Biological Conservation, 144, 383392.
  • Moilanen, A. (2007) Landscape Zonation, benefit functions and target-based planning: Unifying reserve selection strategies. Biological Conservation, 134, 571579.
  • Moilanen, A. & Arponen, A. (2011) Administrative regions in conservation: balancing local priorities with regional to global preferences in spatial planning. Biological Conservation, 144, 17191725.
  • Moilanen, A., Franco, A.M.A., Early, R., Fox, R., Wintle, B. & Thomas, C.D. (2005) Prioritising multiple–use landscapes for conservation: methods for large multi–species planning problems. Philosophical Transactions of the Royal Society of London B, 272, 18851891.
  • Moilanen, A., Anderson, B.J., Eigenbrod, F., Heinemeyer, A., Roy, D.B., Gillings, S., Armsworth, P.R., Gaston, K.J. & Thomas, C.D. (2011a) Balancing alternative land uses in conservation prioritization. Ecological Applications, 21, 14191426.
  • Moilanen, A., Leathwick, J.R. & Quinn, J.M. (2011b) Spatial prioritization of conservation management. Conservation Letters, 4, 383393.
  • Naidoo, R., Balmford, A., Costanza, R., Fisher, B., Green, R.E., Lehner, B., Malcolm, T.R. & Ricketts, T.H. (2006) Global mapping of ecosystem services and conservation priorities. Proceedings of the National Academy of Sciences USA, 105, 94959500.
  • Nelson, E., Mendoza, G., Regetz, J., Polasky, S., Tallis, H., Cameron, D.R., Chan, K.M.A., Daily, G.C., Goldstein, J., Kareiva, P.M., Lonsdorf, E., Naidoo, R., Ricketts, T.H. & Shaw, M.R. (2009) Modeling multiple ecosystem services, biodiversity conservation, commodity production, and tradeoffs at landscape scales. Frontiers in Ecology and the Environment, 7, 411.
  • Normile, D. (2010) UN biodiversity summit yields welcome and unexpected progress. Science, 330, 742743.
  • Orme, C.D.L., Davies, R.G., Olson, V.A., Thomas, G.H., Ding, T.S., Rasmussen, P.C., Ridgely, R.S., Stattersfield, A.J., Bennett, P.M. & Owens, I.P.F. (2006) Global patterns of geographic range size in birds. PLoS Biology, 4, e208.
  • Pajaro, M.G., Mulrennan, M.E. & Vincent, A.C.J. (2010) Toward an integrated marine protected areas policy: connecting the global to the local. Environment Development and Sustainability, 12, 945965.
  • Pressey, R.L. & Bottrill, M.C. (2008) Opportunism, threats, and the evolution of systematic conservation planning. Conservation Biology, 22, 13401345.
  • Pressey, R.L., Cabeza, M., Watts, M.E., Cowling, R.M. & Wilson, K.A. (2007) Conservation planning in a changing world. Trends in Ecology and Evolution, 22, 583592.
  • Ridgely, R.S., Allnutt, T.F., Brooks, T., McNicol, D.K., Mehlman, D.W., Young, B.E. & Zook, J.R. (2007) Digital Distribution Maps of the Birds of the Western Hemisphere, version 3.0. NatureServe, Arlington, Virginia, USA.
  • 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. (2004) Effectiveness of the global protected area network in representing species diversity. Nature, 428, 640643.
  • Rondinini, C., Rodrigues, A.S.L. & Boitani, L.. (2011) The key elements of a comprehensive global mammal conservation strategy. Philosophical Transactions of the Royal Society of London B, 366 (S1) 25912597.
  • Sarkar, S. & Illoldi-Rangel, P. (2010) Systematic conservation planning: an updated protocol. Natureza & Conservacao, 8, 1926.
  • Schuldt, A. & Assmann, T. (2010) Invertebrate diversity and national responsibility for species conservation across Europe – a multi-taxon approach. Biological Conservation, 143, 27472756.
  • Soutullo, A. & Gudynas, E. (2006) How effective is the MERCOSUR's network of protected areas in representing South America's ecoregions? Oryx, 40, 112116.
  • Soutullo, A., De Castro, M. & Urios, V. (2008) Linking political and scientifically derived targets for global biodiversity conservation: implications for the expansion of the global network of protected areas. Diversity and Distributions, 14, 604613.
  • Strange, N., Rahbek, C., Jepsen, J.K. & Lund, M.P. (2006) Using farmland prices to evaluate cost-efficiency of national versus regional reserve selection in Denmark. Biological Conservation, 128, 455466.
  • Tzoulas, K., Korpela, K., Venn, S., Yli-Pelkonen, V., Kazmierczak, A., Niemela, J. & James, P. (2007) Promoting ecosystem and human health in urban areas using Green Infrastructure: a literature review. Landscape Urban Plan, 8, 167178.
  • UNCSD (2012). Available at: http://www.uncsd2012.org/ (accessed 12 June 2012).
  • UNEP/CBD. (2010) Available at: http://www.cbd.int/doc/decisions/COP-10/cop-10-dec-02-en.pdf. (accessed 10 May 2011).
  • Vazquez, L.B., Rodriguez, P. & Arita, H.T. (2008) Conservation planning in a subdivided world. Biodiversity Conservation, 17, 13671377.
  • White, C., Costello, C., Kendall, B.E. & Brown, C.J. (2012) The value of coordinated management of interacting ecosystem services. Ecology Letters, 15, 509519.
  • Wilson, K.A., McBride, M.F., Bode, M. & Possingham, H.P. (2006) Prioritizing global conservation efforts. Nature, 440, 337340.
  • Wilson, K.A., Underwood, E.C., Morrison, S.A., Klausmeyer, K.R., Murdoch, W.W., Reyers, B., Wardell-Johnson, G., Marquet, P.A., Rundel, P.W., McBride, M.F., Pressey, R.L., Bode, M., Hoekstra, J.M., Andelman, S., Looker, M., Rondinini, C., Kareiva, P., Shaw, M.R. & Possingham, H.P. (2007) Conserving biodiversity efficiently: what to do, where, and when. PLoS Biology, 5, 18501861.
  • Wilson, K.A., Evans, M.C., Di Marco, M., Green, D.C., Boitani, L., Possingham, H.P., Chiozza, F. & Rondinini, C.. (2011) Prioritizing conservation investments for mammal species globally. Philosophical Transactions of the Royal Society of London B, 366 (S1) 26702680.

Biosketch

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

Atte Moilanen is a Professor of Conservation Decision Analysis, working at the Finnish Centre of Excellence in Metapopulation Biology, Dept. Biosciences, University of Helsinki. His research group is working on biodiversity conservation informatics, with a focus on development of concepts, methods, analyses and software, primarily for spatial conservation planning, but also for conservation resource allocation in general.

Author contributions: A.M. and C.D.T. conceived the ideas; B.J.A. and A.A. obtained the data; A.A. and B.J. analysed the data; F.M.P. contributed to analysis methods and software development and A.M. and C.D.T. led the writing.

Supporting Information

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

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
ddi12000-sup-0001-AppendixS1-TableS1.docxWord document2201KAppendix S1 Separate analyses for major taxa; mammals, birds, and amphibians. Figure S1 Conservation prioritization for the western hemisphere based on mammals only, corresponding to the all-taxa analysis of the main paper (Fig. 1). Figure S2 Conservation prioritization for the western hemisphere based on birds only, corresponding to the all-taxa analysis of the main paper (Fig. 1). Figure S3 Conservation prioritization for the western hemisphere based on amphibians only, corresponding to the all-taxa analysis of the main paper (Fig. 1). Figure S4 Mean performance curves for per-taxa analyses of Figures S1, S2 and S3; for (a) ABF, (b) CAZ. Table S1 Overlap of the top and bottom 17% of area in the separate analyses for mammals, birds and amphibians.

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.