One of the most frequently cited causes of coral bleaching is anomalously high water temperature (Linden, 1998). The origin of this attribution can be traced to the strong El Niño event of 1982-83, in which widespread bleaching was reported in corals exposed to unusually high surface water temperatures (Glynn, 1988). Since that time, a number of other such observations have been made (Cook et al., 1990; Glynn 1991; Montgomery and Strong, 1994; Brown et al., 1996); and several laboratory studies have demonstrated that elevated seawater temperatures can indeed induce bleaching in corals (Hoegh-Guldberg and Smith, 1989; Jokiel and Coles, 1990; Glynn and D'Croz, 1990).

However, just as anomalously high seawater temperatures have been found to be correlated with coral reef bleaching events, so too have anomalously low seawater temperatures been identified with this phenomenon (Walker et al., 1982; Coles and Fadlallah, 1990; Muscatine et al., 1991; Gates et al., 1992; Saxby et al., 2003; Hoegh-Guldberg and Fine 2004; Yu et al., 2004); and these observations suggest that the crucial link between temperature and coral reef bleaching may not reside in the absolute temperature of the water surrounding the corals, but in the rapidity with which the temperature either rises above or falls below the temperature regime to which the corals are normally adapted. Winter et al. (1998), for example, studied relationships between coral bleaching and nine different temperature indices, concluding that although “prolonged heat stress may be an important precondition for bleaching to occur,” sharp temperature changes act as the “immediate trigger.”

In a related study, Jones (1997) reported coral bleaching on a portion of Australia's Great Barrier Reef just after average daily sea water temperature rose by 2.5°C over the brief period of eight-days. Likewise, Kobluk and Lysenko (1994) observed severe coral bleaching following an 18-hour decline of 3°C in seawater temperature. Because the corals they studied had experienced massive bleaching two years earlier as a result of an anomalous 4°C increase in water temperature, the authors concluded that coral bleaching is more a function of corals not being able to adapt to the rapidity of a temperature change than it is of the absolute magnitude or sign of the change, i.e., heating or cooling.

Further evidence that high or low seawater temperatures per se are not the critical factors in producing coral bleaching is provided by Podesta and Glynn (1997), who examined a number of temperature-related indices of surface waters in the vicinity of Panama over the period 1970-1994. Their analysis revealed that for the two years of highest maximum monthly sea surface temperature, 1972 and 1983, coral bleaching was only reported in 1983, while 1972 produced no bleaching whatsoever, in spite of the fact that water temperatures that year were just as high as they were in 1983.

1.2. Solar Radiation Effects

The link between solar radiation and coral reef bleaching goes back over a century to when MacMunn (1903) postulated that ultraviolet radiation could be potentially damaging to corals. It wasn't until half a century later, however, that scientists began to confirm this suspicion (Catala-Stucki, 1959; Siebeck, 1988; Gleason and Wellington, 1995).

Many investigators of the solar irradiance-coral reef bleaching link have studied the phenomenon by transplanting reef corals from deep to shallow waters. Gleason and Wellington (1993), for example, transplanted samples of the reef-building coral Montastrea annularis from a depth of 24 meters to depths of 18 and 12 meters. Using sheets of acrylic plastic to block out ultraviolet radiation on some of the coral samples, they found that the shielded corals experienced less bleaching than the unshielded corals, and that the unshielded corals at the 12-meter depth had significantly lower amounts of zooxanthellae and chlorophyll per square centimeter than all other treatment and control groups. Likewise, Hoegh-Guldberg and Smith (1989) reported bleaching in the corals Stylophora pistillata and Seriatopora hystrix when they were moved from a depth of 6 meters to 1.2 meters; and Vareschi and Fricke (1986) obtained similar results when moving Plerogyra sinuosa from a depth of 25 meters to 5 meters. As in the case of temperature stress, however, Glynn (1996) notes that artificially reduced light levels have also been observed to cause coral bleaching.

A number of laboratory studies have provided additional evidence for a link between intense solar irradiance and coral reef bleaching; but identifying a specific wavelength or range of wavelengths as the cause of the phenomenon has been a difficult task. Fitt and Warner (1995), for example, reported that the most significant decline in symbiont photosynthesis in the coral Montastrea annularis occurred when it was exposed to ultraviolet and blue light; but other studies have reported coral bleaching to be most severe at shorter ultraviolet wavelengths (Droller et al., 1994; Gleason and Wellington, 1995), while still others have found it to be most strongly expressed at longer photosynthetically-active wavelengths (Lesser and Shick, 1989; Lesser et al., 1990; Brown et al., 1994a).

1.3. Solar Radiation-Temperature Interaction

As additional studies provided evidence for a solar-induced mechanism of coral reef bleaching (Brown et al., 1994b; Williams et al., 1997; Lyons et al., 1998), some also provided evidence for a solar radiation-temperature stress synergism (Gleason and Wellington, 1993; Rowan et al., 1997; Jones et al., 1998). There have been a number of situations, for example, in which corals underwent bleaching when changes in both of these parameters combined to produce particularly stressful conditions (Lesser et al., 1990; Glynn et al., 1992; Brown et al., 1995), such as during periods of low wind velocity and calm seas, which favor the intense heating of shallow waters and concurrent strong penetration of solar radiation.

This two-parameter interaction has much to recommend it as a primary cause of coral bleaching. It is, in fact, the mechanism favored by Hoegh-Guldberg (1999), who claimed – in one of the strongest attempts made to that point in time to portray global warming as the cause of bleaching in corals – that “coral bleaching occurs when the photosynthetic symbionts of corals (zooxanthellae) become increasingly vulnerable to damage by light at higher than normal temperatures.” As we shall see, however, the story is considerably more complicated, more so, even, than what almost everyone has assumed.

1.4. Other Phenomena

In a review of the causes of coral bleaching, Brown (1997) listed (1) elevated seawater temperature, (2) decreased seawater temperature, (3) intense solar radiation, (4) the combination of intense solar radiation and elevated temperature, (5) reduced salinity, and (6) bacterial infections. In a similar review, Meehan and Ostrander (1997) additionally listed (7) increased sedimentation and (8) exposure to toxicants. We have already commented on the four most prominent of these phenomena; and we now address the remaining four.

With respect to seawater salinity, Meehan and Ostrander (1997) noted that, as with temperature, both high and low values have been observed to cause coral bleaching. Low values typically occur as a result of seawater dilution caused by high precipitation events or storm runoff; while high values are much more rare, typically occurring only in the vicinity of desalinization plants.

A number of studies have also clearly delineated the role of bacterial infections in causing coral reef bleaching (Ritchie and Smith, 1998); and this phenomenon, too, may have a connection to high seawater temperatures. In a study of the coral Oculina patagonica and the bacterial agent Vibrio AK-1, for example, Kushmaro et al. (1996, 1997) concluded that bleaching of colonies of this coral along the Mediterranean coast has its origin in bacterial infection, and that warmer temperatures may lower the resistance of the coral to infection and/or increase the virulence of the bacterium. In subsequent studies of the same coral and bacterium, Toren et al. (1998) and Kushmaro et al. (1998) further demonstrated that this high temperature effect may operate by enhancing the ability of the bacterium to adhere to the coral.

In discussing their findings, Kushmaro et al. (1998) commented on the “speculation that increased seawater temperature, resulting from global warming or El Niño events, is the direct cause of coral bleaching.” In contradiction of this presumption, they cited several studies of coral bleaching events that were not associated with any major sea surface temperature anomalies; and they explicitly stated that “it is not yet possible to determine conclusively that bleaching episodes and the consequent damage to reefs is due to global climate change.” Likewise, Toren et al. (1998) noted that the extensive bleaching that occurred on the Great Barrier Reef during the summer of 1982 was also not associated with any major sea surface temperature increase; and they stated that “several authors have reported on the patchy spatial distribution and spreading nature of coral bleaching,” which they correctly noted is inconsistent with the global-warming-induced coral bleaching hypothesis. Instead, they noted that “the progression of observable changes that take place during coral bleaching is reminiscent of that of developing microbial biofilms,” a point that will later be seen to be of great significance.

With respect to sedimentation, high rates have been conclusively demonstrated to lead to coral bleaching (Wesseling et al., 1999); and most historical increases in sedimentation rates are clearly human-induced. Umar et al. (1998), for example, listed such contributing anthropogenic activities as deforestation, agricultural practices, coastal development, construction, mining, drilling, dredging and tourism. Nowlis et al. (1997) also discussed “how land development can increase the risk of severe damage to coral reefs by sediment runoff during storms.” But it has been difficult to determine just how much these phenomena have varied over the last few centuries.

Knowledge in this area took a quantum leap forward, however, with the publication of a study (McCulloch et al., 2003) that provided a 250-year record of sediment transfer to Havannah Reef -- a site on the inner Great Barrier Reef of northern Queensland, Australia -- by flood plumes from the Burdekin River. According to the authors of that study, sediments suspended in the Burdekin River contain barium (Ba), which is desorbed from the particles that carry it as they enter the ocean, where growing corals incorporate it into their skeletons along with calcium (Ca). Hence, when more sediments are carried to the sea by periodic flooding and more gradual longer-term changes in land use that lead to enhanced soil erosion, the resultant increases in sediment load are recorded in the Ba/Ca ratio of coral skeleton material. Inspired by these facts, McCulloch et al. measured Ba/Ca ratios in a 5.3-meter-long coral core from Havannah Reef that covered the period from about 1750 to 1985, as well as in some shorter cores from Havannah Reef and nearby Pandora Reef that extended the proxy sediment record to 1998.

Results of the analysis revealed that prior to the time of European settlement, which began in the Burdekin catchment in 1862, there was "surprisingly little evidence for flood-plume related activity from the coral Ba/Ca ratios." Soon after, however, land clearance and domestic grazing intensified and the soil became more vulnerable to monsoon-rain-induced erosion. By 1870, baseline Ba/Ca ratios had risen by 30% and "within one to two decades after the arrival of European settlers in northern Queensland, there were already massive impacts on the river catchments that were being transmitted to the waters of the inner Great Barrier Reef." During subsequent periods of flooding, in fact, the transport of suspended sediment to the reef increased by fully five to ten-fold over what had been characteristic of pre-European settlement times.

In a companion article, Cole (2003) reported that corals from East Africa "tell a similar tale of erosion exacerbated by the imposition of colonial agricultural practices in the early decades of the twentieth century." There, similar coral data from Malindi Reef, Kenya, indicate "a low and stable level of barium before about 1910 which rises dramatically by 1920, with a simultaneous increase in variance," a phenomenon that was also evident in the Australian data.

What are the implications of these observations? Cole concludes that "human activity, in the form of changing land use, has added sedimentation to the list of stresses experienced by reefs." Furthermore, as land-use intensification is a widespread phenomenon, she notes that "many reefs close to continents or large islands are likely to have experienced increased delivery of sediment over the past century," which suggests that the stress levels produced by this phenomenon are likely to have increased over the past century as well. In addition, Cole logically concludes that as coastal populations continue to rise, "this phenomenon is likely to expand."

Taken together, these findings suggest that the natural course of human population growth and societal and economic development over the period of the Industrial Revolution may have predisposed coral reefs to ever-increasing incidences of bleaching and subsequent mortality via a gradual intensification of near-coastal riverine sediment transport rates.

Lastly, a number of poisonous substances are known to have the capacity to induce coral bleaching. Some of them are of human origin, such as herbicides, pesticides and even excess nutrients that ultimately make their way from farmlands to the sea (Simkiss, 1964; Pittock, 1999). Other poisons originate in the sea itself, many the result of metabolic waste products of other creatures (Crossland and Barnes, 1974) and some a by-product of the coral host itself (Yonge, 1968). Each of these toxicants presents the coral community with its own distinct challenge.