The small island states account for less than 1% of global GHG emissions but are among the most vulnerable of all areas to the potential adverse effects of climate change and sea-level rise (Jones, 1998; Nurse et al., 1998). It has been established that there already is a global commitment to climate change and sea-level rise as a result of greenhouse forcing arising from historic emissions (Warwick et al., 1996; Jones, 1998; Nicholls et al., 1999; Parry et al., 1999). Moreover, analysis has shown that even with a fully implemented Kyoto Protocol, by 2050 warming would be only about 1/20th of a degree less than what is projected by the IPCC (Parry et al., 1999). Therefore, climate change impacts are inevitable.
Thus, owing to their high vulnerability and low adaptive capacity to climate change, communities in small island states have legitimate concerns about their future on the basis of the past observational record and present climate model projections. Economic development, quality of life, and alleviation of poverty presently constitute the most pressing concerns of many small island states. Thus, with limited resources and low adaptive capacity, these islands face the considerable challenge of charting development paths that are sustainable and controlling GHG emissions, without jeopardizing prospects for economic development and improvements in human welfare (Munasinghe, 2000; Toth, 2000). At the same time, given the inevitability of climate change and sea-level rise, they are forced to find resources to implement strategies to adapt to increasing threats resulting from GHG forcing of the climate system, to which they contribute little (Hay and Sem, 1999; Sachs, 2000). Consequently, the already meager resources of these island states will be placed under further pressure.
Figure 17-6: Climate change scenarios for Mediterranean Sea islands as simulated by five AOGCMs for the 2020s, 2050s, and 2080s.
Although the severity of the threat will vary regionally, sea-level rise of the magnitude currently projected (i.e., 5 mm yr-1, with a range of 2-9 mm yr-1), is expected to have disproportionately great effects on the economic and social development of many small island states (Granger, 1997; IPCC, 1998). Coastal land loss already is projected to have widespread adverse consequences. Indeed, it is argued that land loss from sea-level rise, especially on atolls (e.g., those in the Pacific and Indian Oceans) and low limestone islands (e.g., those in the Caribbean), is likely to be of a magnitude that would disrupt virtually all economic and social sectors in these countries (Leatherman, 1997). Recent estimates indicate that with a 1-m rise in sea level, 10.3 km2 of land in Tongatapu island, Tonga, would be lost (Mimura and Pelesikoti, 1997). This figure would increase to 37.3 km2 (14%) with storm surge superimposed on a 1-m sea-level rise scenario. For some main Yap Island (Federated States of Micronesia) sites, a retreat of 9 to 96 m is projected with a 1-m rise in sea level (Richmond et al., 1997). On Majuro Atoll, Marshall Islands, land loss from one area based on the Bruun rule is estimated to be nearly 65 ha of dry land from a 1-m rise in sea level (Holthus et al., 1992).
One of the most serious considerations for some small islands is whether they will have adequate potential to adapt to sea-level rise within their own national boundaries (Nurse, 1992; IPCC, 1998). In tiny islands where physical space already is very scarce, adaptation measures such as retreat to higher ground and use of building set-backs appear to have little practical utility. In extreme circumstances, sea-level rise and its associated consequences could trigger abandonment and significant "off-island migration," at great economic and social costs (Leatherman, 1997; Nicholls and Mimura, 1998).
Box 17-1. Climate Change Scenarios for the South Pacific Region
As part of the Pacific Islands Climate Change Assistance Programme, the Commonwealth Scientific and Industrial Research Organisation (CSIRO) has prepared regional climate change scenarios for the four South Pacific regions of Micronesia, Melanesia, and north and south Polynesia (Jones et al., 1999). Six coupled atmosphere-ocean climate simulations were included in the analysis of regional climate change scenarios: CSIRO Mark 2 GCM with and without sulfates, CSIRO DARLAM 125 km, DKRZ ECHAM4/OPYC3 GCM, Hadley Centre HADCM2, and the Canadian CGCM1. The regional scenarios derived can be considered as projections that represent a range of possible future climates.
Generally, the models project a temperature increase that is less than the global mean. Results show the least warming in the South Pacific; regional maximum warming is projected in the far west, central, and eastern equatorial Pacific. Four of the models show an increase in rainfall over the central and eastern Pacific over both half-years (i.e., May to October and November to April). Movements of both the ITCZ and the SPCZ were not consistent between models, but rainfall consistently increased. Increases in daily rainfall intensity are expected in regions where rainfall increases, remains the same, or decreases slightly, as derived from several models and studies. Thus, high confidence is attached to this result.
Historical sea-level rise over the Pacific from tide gauge records adjusted for postglacial rebound is consistent with global estimates of 1-2 mm yr-1. ENSO is the dominant influence on climate variability in the Pacific, and model outputs show that the ENSO phenomenon is likely to continue to 2100. The results also suggest that under climate change, there is likely to be a more El Niño-like mean state over the Pacific. There is no evidence that tropical cyclone numbers may change, but a general increase in tropical cyclone intensity, expressed as possible increases in wind speed and central pressure of 10-20% with 2xCO2 equivalent, now appears likely. Moderate confidence is attached to this result. No significant change in regions of formation was noted in the DARLAM 125 km resolution simulation, although this may alter in response to long-term changes to ENSO. Although there appears to be no major change in regions of origin, tropical cyclones showed a tendency to track further poleward. Low confidence is attached to this result.
Changes in the highest sea levels at a given locality will result from the change in mean sea level at that location and changes in storm-surge heights. If mean sea level rises, present extreme levels will be attained more frequently, all else being equal. The increase in maximum heights will be equal to the change in the mean, which implies a significant increase in the area threatened with inundation. This will be especially true in areas with a small surge envelope, which is typical in most small islands. Under such circumstances, even incrementally small elevations in sea level would have severely negative effects on atolls and low islands (Forbes and Solomon, 1997; Nicholls et al., 1999).
Changes in storm-surge heights also would result from alterations in the occurrence of strong winds and low pressures, as would occur during the passage of tropical storms and cyclones. It is already known that tropical cyclones are the major cause of storm surges that impact small islands in the Atlantic, Pacific, and Indian Oceans. Changes in the frequency and intensity of tropical cyclones could result from alterations to SST, large-scale atmospheric circulation, and the characteristics of ENSO (Pittock et al., 1996). Although there is no consensus yet with regard to whether there will be changes in the behavior of these systems (Royer et al., 1998; Jones et al., 1999), the prospect of extreme sea levels (related to storm surges and higher wave amplitudes) is a concern that small island states cannot easily ignore.
Based on global sea-level rise scenarios produced by the Hadley Centre (HADCM2 and HADCM3), Nicholls et al. (1999) estimate that global sea levels are expected to rise by about 38 cm between 1990 and the 2080s. They project that many coastal areas are likely to experience annual or more frequent flooding, with the islands of the Caribbean and the Indian and Pacific Oceans facing the largest relative increase in flood risk. Projected out to the 2080s, the number of people facing high flood risk from sea-level rise in these regions would be 200 times higher than in the case of no climate change (Nicholls et al., 1999). Recent studies for Cuba (based on HADCM2 and IS92a scenarios) also project that 98 coastal settlements with a combined population exceeding 50,000 persons would be inundated by a 1-m rise in sea level (Perez et al., 1999).
Other reports in this collection