species can settle, whereas without disturbance, superior competitors drive out inferior competitors in a biological succession that ends in lowered diversity. If this theory were used to try to explain diversity gradients, it might suggest that the area of highest diversity is where disturbances are of intermediate intensity and frequency. However, sea surface temperatures in Indonesia and New Guinea are disturbed less by cold events than in areas farther from the equator. Similarly, there are no cyclones near the equator in Indonesia (including Papua) and New Guinea (Figure 5.2.7a,b; Fenner and Riolo, under review). On the other hand, the northern Philippines experiences a moderate to high level of disturbance from cyclones, and yet is part of the Coral Triangle. Cyclones are probably one of the most important natural disturbances on coral reefs (Rogers 1993). Thus, the Intermediate Disturbance Hypothesis is unlikely to explain coral diversity gradients.
In ecosystems with low diversity, each species may be represented by large numbers of individuals. For example, in the tundra of northern Canada and Alaska, there are millions of Snow Geese (Chen hyperboreus) in the summer, and millions of mosquitoes. There is only one large mammal, the Caribou (Rangifer caribou) and it is also present in large numbers. Northern forests are composed of large numbers of individuals of a small number of tree species. By contrast, in the tropics there are large numbers of species, most of which are rare. In a hectare of rainforest in New Guinea, there can be several hundred species of trees, but few individuals of each species. An example from coral reefs is sea slugs (opisthobranchs): in the western Pacific there are many species (over 500), most of which are very rare. In the tropics, functional groups of species, called guilds, usually have many more species than in the temperate or, especially, polar areas. Research on terrestrial systems indicates that the loss of individual species does not have a great impact on a high-diversity ecosystem, because there are many other members of most guilds that can continue to perform that guild’s functions (Grime 1997; Moffat 1996). The loss of a member of a guild in a low-diversity ecosystem may have a much larger impact on the ecosystem, particularly if there is only one member of that guild so that with its loss the guild function is no longer performed. Similarly, Bellwood et al. (2004) have argued that low diversity coral reefs are more vulnerable to the loss of individual species than diverse ecosystems for these reasons. For example, the loss of a single species of sea urchin, Diadema antillarum, in the Caribbean in 1983–1984 (Lessios 1988; Lessios, Robertson, Cubit 1984), led to major phase shifts on some reefs from coral-dominated reefs to algal-dominated hard grounds. The large numbers of a single species in low-diversity ecosystems also makes them more vulnerable to diseases and specialized predators. A dense population of a single species, as was the case with D. antillarum, makes the transmission of disease easier. The die-off of D. antillarum was the largest marine epizootic ever recorded. Similarly, two of the most common coral species in the Caribbean were Acropora palmata and A. cervicornis. Both form large, dense, single-species thickets of genetically identical organisms, or clones. The lack of genetic diversity means that any disease that can kill one individual can kill the whole clone. Both of these species have been decimated in much of the Caribbean by White Band disease (Aronson, Precht, and Macintyre 1998), and were considered for Endangered Species status (Precht, Robbart, and Aronson 2004; Shinn 2004; Wilkinson 2004); they received protected status on 8 June 2006. On diverse coral reefs, most species are rare, so disease transmission is much more difficult, and the likelihood of epizootics is reduced. This probably contributes to the stability of high diversity coral reefs.
Figure 5.2.7. a. Tracks of tropical cyclone and severe storm tracks for the Indo-Pacific region. All storms that were classified above a tropical depression in strength (wind speeds > 30 mph for two or more 6-hour periods) from 1945 to 2003 are included. b. Severe storm density in the Indo-Pacific region. Storm density was computed for each 50 50 km cell by summing the number of tracks found within 200 km of the cell and dividing by the total area sampled in each cell.
Source: Data assembled by Unisys Corporation and Joint Typhoon Warning Center (www.npmoc.navy.mil/jtwc.html) and downloaded from the Pacific Disaster Center website (atlas.pdc.org). Maps by F. Riolo.
Effects of Fishing
Fishing can remove fish that are important for reef health. The removal of herbivorous fish in Jamaica left it vulnerable, so when a disease killed the last herbivore (sea urchins), the reef was overcome with algae (Hughes et al. 1987). Large fish such as sharks, Humphead Wrasse (Cheilinus undulatus), and Bumphead Parrot-fish (Bulbometopon muricatum) are particularly vulnerable. Bumphead Parrotfish have been extirpated from several places in the Indo-Pacific (Bellwood et al. 2003; Dulvy et al. 2003). Humphead Wrasses are under heavy pressure over a large area due to the live food fish trade. Populations are reduced to low levels in areas with higher fishing pressure (Sadovy et al. 2003). Areas of Fiji where fishing pressure is greatest are also the areas where Crown-of-Thorns starfish have outbreaks and eat the tissue off corals, killing the corals (Dulvy et al. 2004). Humphead Wrasses are known to eat toxic invertebrates like the Crown-of-Thorns starfish, so overfishing them may leave reefs vulnerable to Crown-of-Thorns attacks. Although the reefs of Papua are remote and under relatively little pressure from human populations (but see Birkeland 1982), fishing makes sharks and Humphead Wrasse rare, with just two adult Humphead Wrasse seen in 45 sites (Allen 2002b; McKenna et al. 2002).
Few coral reef researchers or managers have seen what truly pristine coral reef fish populations are like, and the amazing dominance of apex predators. Each generation of scientists remembers what coral reefs were like when they first saw them, and tend to think of that as the standard of undisturbed ecosystems. We have lowered our standards in a process called ‘‘shifting baselines.’’ Coral reefs in the Caribbean have declined in three decades from about 50% coral cover to about 10% coral cover (Gardner et al. 2003). Archeological methods have been used to study the effects of pre-Columbian fishing in the Caribbean, and the studies have found that declines began even before the arrival of Europeans (Wing and Wing 2001). Paleontological methods along with archeological and historical methods have shown declines in 14 coral reef systems worldwide since pre-human times (Pandolfi et al. 2003). The 14 reef systems studied had declined between about 28% and 78% of the way from pristine toward ecologically extinct.
Endemism and Extinction
Endemism is commonly used in terrestrial conservation programs as a measure of the need to conserve areas (Allen 2003). It is especially important to avoid the local extinction of endemic species, because their loss represents the loss of an entire species. Endemic species are more vulnerable to extinction partly because any local disturbance can cause global extinction, and also because endemic species usually have small populations. The rates of endemism on coral reefs are quite different in different groups of organisms. Endemism is uncommon in larger organisms, but may be high in some groups of small organisms, and low in the tiniest microscopic organisms. Most groups of larger coral reef organisms have wide dispersal and very few species are endemic. Many coral reef species are broadcast spawners, releasing tiny eggs into the water that are carried with the currents. Currents can carry the eggs considerable distances during the several days to weeks required for them develop to the stage where they are ready to settle. For example, one species of sea urchin, Echinothrix diadema, has been found to be genetically the same species in Hawai’i and the east Pacific, across the largest expanse of open water in the tropics anywhere in the world (Lessios et al.1998). Some coral species have been observed attached to floating objects and thus are probably able to ‘‘raft’’ over vast distances (Jokiel 1990). This is even true of species that brood their young, releasing larvae that quickly attach close to the parent. Reef fish have also been observed rafting by staying near floating debris (Mora 2001). This wide dispersal means that there are few endemic species on most coral reefs, especially in the western Pacific where there are many reefs close together. For instance, currently no endemic coral species are known in the Philippines or Indonesia, where 535 and 581 coral species are currently known, respectively (Fenner, under review c; Turak 2003;
