Report

Marine Biodiversity Hotspots and Conservation Priorities for Tropical Reefs

See allHide authors and affiliations

Science  15 Feb 2002:
Vol. 295, Issue 5558, pp. 1280-1284
DOI: 10.1126/science.1067728

Abstract

Coral reefs are the most biologically diverse of shallow water marine ecosystems but are being degraded worldwide by human activities and climate warming. Analyses of the geographic ranges of 3235 species of reef fish, corals, snails, and lobsters revealed that between 7.2% and 53.6% of each taxon have highly restricted ranges, rendering them vulnerable to extinction. Restricted-range species are clustered into centers of endemism, like those described for terrestrial taxa. The 10 richest centers of endemism cover 15.8% of the world's coral reefs (0.012% of the oceans) but include between 44.8 and 54.2% of the restricted-range species. Many occur in regions where reefs are being severely affected by people, potentially leading to numerous extinctions. Threatened centers of endemism are major biodiversity hotspots, and conservation efforts targeted toward them could help avert the loss of tropical reef biodiversity.

Coral reefs fringe one-sixth of the world's coastlines (1) and support hundreds of thousands of animal and plant species (2). Fifty-eight percent of the world's reefs are reported to be threatened by human activities (3). Terrestrial agriculture, deforestation, and development are introducing large quantities of sediment, nutrients, and other pollutants into coastal waters, causing widespread eutrophication and degradation of biologically productive habitats (4, 5). Coral reefs are often fished intensively; and in regions of the Indian and Pacific Oceans, fishing with dynamite and poisons has devastated reef habitats (6). Coral reefs are also susceptible to climate change; 25% of the world's coral reefs have already been destroyed or severely degraded through problems arising from climate warming (7). Among marine ecosystems, tropical reefs represent a high priority for conservation action [Web note 1 (8)].

We used data on the distribution of 3235 species from four phyla to explore the potential consequences of widespread reef degradation for biodiversity and to investigate ways to target conservation action to places where it is most needed and could have the greatest benefits. We mapped the geographic ranges of 1700 species of reef fish, 804 species of coral, 662 species of snail, and 69 species of lobster. We chose these taxa because they are well-known, good distributional data for them are available on a global scale, and they represent reasonable surrogates for reef diversity as a whole (9) [Web note 2 and Web table 1 (8)]. Figure 1, A through D, shows global clines in species richness of these taxa, mapped on an equal-area grid (10). There is a high level of concordance in patterns of total species richness across the four taxa (pairwise Spearman's rank correlations range from 0.78 to 0.89). For all taxa, species richness peaks in the so-called “coral triangle” of Southeast Asia (11), then falls off rapidly moving east across the Pacific, and less rapidly moving west across the Indian Ocean. In the tropical Atlantic, all taxa have highest richness in the Caribbean.

Figure 1

Global clines in species richness of fish (A), corals (B), snails (C), and lobsters (D). Scales show number of species present. (E) Concordance of the top 10% most species-rich cells among taxa. Red cells were included for all four taxa, orange for three, yellow for two, and blue for one. (F) Threats to reefs in each grid cell, calculated using data from Bryant et al. (3, 13). Blue represents low risk (ave- rage threat score between 1 and 1.67); yellow, medium risk (score between 1.68 and 2.33); and red, high risk (score ≥2.34). (G) Concordance in patterns of range rarity among the top-scoring 10% of cells for each taxon. Color codes are as in (E). Places outlined show multitaxon centers of endemism (13) [Web table 2 (8)], numbered as in Table 2.

Figure 1E shows a high degree of overlap in the top 10% most species-rich cells for each taxon. 26.5% of the richest cells were shared by four taxa, 38.6% by three, and 38.6% by two (12). Cells in the southern Philippines and central Indonesia are in the top 10% richest locations for all four taxa, and degree of overlap declines moving away from this region.

Figure 1F shows the distribution of threats to coral reefs from human impacts, based on an analysis by Bryant et al.(3). They mapped threats to reefs from coastal development, overexploitation, and pollution from marine and land-based sources, then classified reefs as facing low, medium, and high levels of threat. Using their data, we calculated the average threat to reefs in each grid cell on a scale of 1 to 3 (low to high threat) (13). Areas of greatest species richness are exposed to significantly greater threats from human impacts than are less rich regions (Table 1).

Table 1

Comparisons of threats to reefs from human impacts in the top 10% highest scoring cells versus the bottom 90% of cells. Threat scores for each grid cell were calculated with data from Bryantet al. (3, 12) and range from 1 (low threat) to 3 (high threat). Figures show mean scores. For all taxa, the top 10% most species-rich cells were significantly more threatened than the rest. For the top 10% of range-rarity scores, this was the case only for corals and snails.

View this table:

Marine species have long been considered resilient to extinction because of their large geographic range sizes (14). Our data contradict this view for three of the four taxa. Figure 2 shows cumulative curves for species number versus range size, expressed as the number of cells within a species' extent of occurrence that contained reef habitat. Although most corals are widespread, most lobsters are geographically restricted, and fish and snails have roughly equal numbers of restricted-range and widespread species. Even among corals, 58 species (7.2%) had restricted ranges (≤10 cells). The figures were 26.5, 28.7, and 53.6%, respectively, for fish, snails, and lobsters. Hence, restricted-range species are common in the sea, and widespread reef degradation could lead to a gathering wave of extinctions. The low fraction of restricted-range corals should be treated with caution because we identified species by morphology (15). Corals and many other marine organisms with similar morphology conceal substantial genetic differences, even across regions without obvious barriers to gene flow (16, 17). Future studies may reveal much cryptic speciation that could revise our conclusion that coral species are generally widespread. Extinction risk could be greater than suggested by our findings.

Figure 2

Cumulative curves of species number versus geographic range size. Range size is expressed as the number of cells containing reef habitat within each species' extent of occurrence.

We examined the distribution of geographically restricted species to determine whether such endemics are clustered together into centers of endemism as they are in terrestrial environments (18). We used the reciprocal of the range size of each species as a measure of range rarity, and for each cell we summed the values of all species present (19). Maps of range-rarity scores reveal areas that are rich in restricted-range species [Web fig. 1, A through D (8)]. We mapped the top-scoring 10% of cells for each taxon, and Fig. 1G shows that there is high concordance among them. Of the top-scoring cells, 54.2% were shared by two taxa, 22.9% by three, and 3.6% by four (12).

Centers of endemism predominate in places isolated by distance or oceanography. For example, isolated islands rich in endemics include Mauritius and La Reunion in the Indian Ocean, Hawaii and Easter Islands in the Pacific, and St. Helena and Ascension Islands in the Atlantic. Centers of endemism also occur where nonreversing currents move water from tropical to temperate latitudes. Examples include east and west Australia, eastern South Africa, and southern Japan. However, there are also multitaxa centers of endemism in places that appear highly interconnected with other regions, such as the Philippines, Sunda Islands, and New Caledonia. This accords with growing evidence that species with pelagic larval stages do not always disperse widely (20–23).

We identified the 18 richest multitaxon centers of endemism (Fig. 1G) (24) [Web table 2 (8)]. They include 35.2% of the world's coral reefs and cover only 0.028% of the world's oceans (25), but include between 58.6 and 68.7% of restricted-range species from the four taxa (ranges ≤10 grid cells). The 10 richest centers of endemism cover just 15.8% of the world's coral reefs, but include between 44.8 and 54.2% of restricted-range species.

Terrestrial biodiversity hotspots have been defined on the basis of both endemism and threats facing them [the loss of >70% of primary vegetation (18)]. Figures for loss of primary habitat are unavailable for tropical reefs, but we can examine the risks of habitat loss based on threats estimated in Bryant et al.'s assessment (3, 13,26). Many centers of endemism are deeply at risk and can be considered analogous to terrestrial biodiversity hotspots (Fig. 3). Without rapid conservation action, species will be lost. We define 10 marine biodiversity hotspots as those centers of endemism with average threat scores above 1.67 (that is, in the top two-thirds of the range of risk from human impacts) (Table 2 and Fig. 3). Focusing conservation effort on them could be highly effective in preventing species loss (27), but how good would it be as a strategy for protecting more widespread species?

Figure 3

Threats to reefs in centers of endemism. The figure shows mean (circles), maximum, and minimum threat scores for grid cells included within each center of endemism, calculated with data from Bryant et al. (3,13).

Table 2

Summary of attributes of centers of endemism [see Web table 2 (8) for details of places included in each]. We define marine biodiversity hotspots, indicated by bold type, as centers of endemism with average threat scores above 1.67 (3,13). na, not available.

View this table:

Measures of range rarity and species richness are closely coupled for corals [Spearman's rank correlation (SR) = 0.86,P < 0.001, n = 825], loosely coupled for snails (SR = 0.51, P < 0.001), and largely uncoupled for fish (SR = 0.12, P < 0.001) and lobsters [SR = 0.06, not significant (NS)]. This means that although targeting centers of endemism for conservation would also benefit broader elements of coral diversity, it is likely to be less effective for other taxa. However, from the perspective of species' representation, the strategy looks better. The 10 richest centers of endemism include representatives of between 59.4 and 75.2% of all species in our sample, depending on taxon, whereas all 18 include from 73.9 to 96.1%.

Figure 3 shows that even the most deeply threatened centers of endemism often include places where threats are relatively low. Timely investment in the protection of these areas could yield good results. However, adopting a two-pronged conservation strategy, as Myerset al. (18) suggested for terrestrial ecosystems, would be better. Extensive areas of coral reef remain little affected by people, yet are rich in species (Fig. 1, A through F). Conservation efforts should extend to both marine biodiversity hotspots and reef “wilderness” areas and must include efforts to mitigate climate change.

Many threats to tropical reefs originate on land, including downstream impacts of forest loss, agricultural expansion, and construction (3). Our analysis reveals an opportunity for integrating terrestrial and marine conservation. Eight of 10 marine biodiversity hotspots and 14 of 18 centers of endemism are adjacent to terrestrial biodiversity hotspots (18) (Table 2). Extending terrestrial conservation efforts seaward in those places offers an effective and affordable strategy for protecting the planetary biota (18, 27, 28).

  • * To whom correspondence should be addressed at the Environment Department, University of York, York, YO10 5DD, UK. E-mail: cr10{at}york.ac.uk

  • Deceased.

REFERENCES AND NOTES

View Abstract

Navigate This Article