As a result of global warming, the penetration of heat into the ocean leads to the thermal expansion of the water; this effect, coupled with the melting of glaciers and ice sheets, results in a rise in sea level. Sea-level rise will not be uniform globally but will vary with factors such as currents, winds, and tides-as well as with different rates of warming, the efficiency of ocean circulation, and regional and local atmospheric (e.g., tectonic and pressure) effects.
For the IS92a emission scenario, it is estimated that sea level would rise, on average, about 5 mm/yr, within a range of uncertainty of 2-9 mm/yr (IPCC 1996, WG I, Section 126.96.36.199). On regional and local scales, sea level is expected to vary significantly from this global projection, as a result of factors such as vertical land movements (Aubrey and Emery, 1993; Hendry, 1993; Maul, 1996) and dynamic effects due to ocean circulation, wind and pressure patterns, and ocean-water density.
Although these projections are lower than the IPCC (1990) estimates, it should be emphasized that the current best estimates for sea-level rise represent a rate approximately two to four times higher than the rate experienced in the past 100 years (i.e., 1.0-2.5 mm/yr). Model runs also show that sea level would continue to rise beyond the year 2100 (because of lags in the climate response), even with assumed stabilization of global GHG emissions (IPCC 1996, WG II, Section 188.8.131.52). Recent estimates by Pittock et al. (1995) indicate a sea-level rise of 28-32 cm for the southWest Pacific region at the time of CO2 doubling-the figure representing only the contribution from thermal expansion. These projections are of considerable concern to small islands (Pernetta, 1992; Hay and Kaluwin, 1993; Maul, 1993): Many islands and atolls in the Pacific and Indian Oceans rarely exceed 3-4 m above mean sea level in elevation and therefore could be vulnerable to changes of this magnitude. The extent of this threat, however, would depend on several factors, including the rate of sand production.
The ecological systems of small islands-and the functions they perform-are sensitive to the rate and the magnitude of changes in climate. These systems provide food, medicine, and energy; process and store carbon and other nutrients; assimilate wastes; purify water and regulate runoff; and provide opportunities for recreation and tourism (see IPCC 1996, WG II, Section 9.2).
Although GCMs provide little insight at present into the extent of climate change impacts on island ecosystems, it is widely postulated that some effects could be quite dislocating for the inhabitants of many small islands (Pernetta, 1988; Roy and Connell, 1991; Holthus et al., 1992; Maul, 1993, 1996; Nicholls, 1995). Various socioeconomic sectors, including tourism, infrastructure, agriculture, water resources, and human health-all of which are sensitive to fluctuations in rainfall, temperature, and sea level-also could be negatively affected.
Coral reefs represent one of the most important resources of tropical islands. They perform valuable functions, including supplying sand to beaches and playing a critical role in the formation and maintenance of reef islands; they are habitats for a variety of marine communities; and they serve as spawning and nursery grounds for numerous species of reef fish. Reefs also function as protective barriers for beaches and coasts by reducing incident wave energy through the processes of wave reflection, dissipation, and shoaling; they also are significant contributors to the economic resource base of many small island states.
Given current projected rates of increase, sea-level rise per se is not expected to have widespread adverse effects on coral reefs. Indeed, some researchers argue that a rising sea level actually may be beneficial because the new conditions would be favorable for inducing vertical growth; in contrast, reef growth has been largely horizontal in the recent past, as a consequence of lower sea levels (Wilkinson and Buddemeier, 1994). Edwards (1995) further suggests that even slowly accreting reef flats should be able to cope with projected sea-level rise, in the absence of other negative forces (e.g., elevated seawater temperature and anthropogenic stresses).
The climate change effect of greatest potential significance to coral reefs is likely to be an increase in seawater temperature. Corals have narrow temperature tolerances (approximately 25-29�C) and salinity tolerances (about 32-36 ppm) (Bellairs Research Institute, 1990). In some islands, some species of corals currently live at or near their threshold of temperature tolerance (Goreau, 1992). Corals respond to the combined effects of irradiance and water temperature elevation by paling in color, or bleaching. However, corals generally do not bleach simply as a result of rapid fluctuation in water temperature but rather as a result of departures in temperature above their seasonal maximums. If the temperature elevation is substantial over an extended period (e.g., 3-4�C for >6 months), significant coral mortality is likely (Brown and Suharsono, 1990). On the other hand, if the temperature increase is relatively small (e.g., 1-2�C) for a short period, bleached corals may recover, though with reduced growth and impaired reproductive capabilities (Goreau, 1992; Brown and Ogden, 1993). A comprehensive review of the causes and consequences of coral bleaching has recently been completed (Brown, 1997).
Notwithstanding their adaptive capacity, corals-like other biotic communities-will continue to be subjected to increasing human stresses (e.g., nutrient loading and other types of chemical pollution, sedimentation from land-based activities, damage from anchoring of boats). These pressures inevitably will limit the innate capacities of these organisms to adapt to the effects of climate change.
ENSO events already have been associated with extensive coral bleaching in the Caribbean and the Pacific in the early 1990s. Although it is not yet clear how climate change will influence the incidence of ENSO events in the future, events that cause high temperature excursions will lead to coral bleaching and, possibly, to mortality (Goreau, 1992; Wilkinson and Buddemeier, 1994; see also Glynn, 1993).
Other reports in this collection