Posted March 1998
Potential impacts of sea-level rise
on populations and agriculture
by R. Gommes, J. du Guerny, F. Nachtergaele and R. Brinkman
Food and Agriculture Organization of the United Nations
Cover page | 1. Introduction / setting | 2. Effects of sea-level rise | 3. Lessons from the past | 4. Vulnerability index | 5. Conclusions, references | 6. Appendix table: AOSIS countries
According to Nicholls and Leatherman (1995), a 1m sea-level rise would affect 6 million people in Egypt, with 12% to 15% of agricultural land lost, 13 million in Bangladesh, with 16% of national rice production lost, and 72 million in China and "tens of thousands" of hectares of agricultural land.
More than direct land loss due to seas rising, indirect factors are generally listed as the main difficulties associated with sea-level rise. These include erosion patterns and damage to coastal infrastructure, salinization of wells, sub-optimal functioning of the sewerage systems of coastal cities with resulting health impacts (WHO 1996, chapter 7), loss of littoral ecosystems and loss of biotic resources.
Deltas pose a set of more specific problems, as they are the areas where sea and land most closely interact: their average elevation is usually very low, to the extent that tidal effects can be felt for several tens of km, and in some cases hundreds, inland. Note that the deltas in closed seas, such as the Nile and Danube, do not normally suffer the same difficulties. This land/sea interaction results in very complex agricultural systems, where irrigation and rain-fed agriculture may be practised in alternate seasons, with attention to irrigation water quality (salinity) and to the washing out of salts by rains before planting crops.
The Mekong delta provides a clear example of this complexity (see, for example, Jelgersma et al., 1993). Siltation is usually very active during certain seasons of the year: deposition in the Mekong Delta continues to extend the Ca Mau peninsula to the south and west at a rate of 150 m/annum in some places (Fedra et al, 1991). The delta, therefore, is as much the result of conditions up-stream as the result of local or coastal interactions. Sea-level rise, therefore, cannot be examined without some sound assumptions about climate and rainfall changes in the river catchment as a whole.
Contrary to many other "normal" coasts, deltas are difficult and expensive to protect, due to the very dissected nature of the coastline. A measure of the difficulty is provided by the "multiplier"  used by CZMS (1990) to evaluate protection costs. Multipliers for deltas usually vary from 6 to 8, but may reach higher values for the Magdalena (10.4, Venezuela), Orinoco (11.3), Parana (26.7), Mekong (9.8) or Ganges-Brahmaputra-Mahanadi (7.5) CZMS (1990) stresses that the multiplier may be affected by large errors (100%), but the parameter definitely gives a measure of the level of intricacy of land and sea in deltas.
Although deltas tend to be densely populated, it is not really relevant to assess exactly how many people live in them . Due to their very high productivity (generally fertile soils, water availability, multiple cropping, especially in tropical areas) they produce significantly more food than the local consumption. In Vietnam, for instance, 50% of national rice production comes from the Mekong delta in the south, while 20% is produced in the Red River delta near Hanoi. Although much of this production is for export, a disaster in the deltas would have profound effects on the whole country. This indicates a fragile situation in which any major disturbance would result in economic and possibly political shock waves well beyond the delta proper.
Negative values indicate rainier conditions along the coasts, defined a 0.5 square (up to 60 km from the coast at the equator), while land areas are all land found at the same latitude. Based on the IIASA data set of Leemans and Cramer, 1990. The graph shows 5 degree moving-averages.
While it is possible, at any location, to define a "climate complex", i.e. a set of inter-related climatic variables (Sombroek and Gommes, 1996), it should be noted in the current context that coastal climate complexes tend to be rather different from the climate complexes at some distance from the sea (Figure 4 above). Since delta agriculture typically uses a combination of irrigated and rain-fed crops, a change in the relative behaviour of the local delta climate and the climate in the basin may lead to unprecedented difficulties. It might impose, for instance, a shift to more salt tolerant crops.
It is unknown how global climate change would alter these patterns, but is it obvious that zonal scenarios are unlikely to apply on the coast. Similarly, the sea- and land breeze patterns are very likely to be altered.
In coastal areas, and particularly deltas, we thus have to take into account: projected sea-level rise, modified ocean circulation patterns as they affect the building and erosion of the coast, climate change in the catchment basin and change of the coastal climate, and changes in the frequency of extreme events. This also constitutes a major difference with small islands where remote land-based changes are likely to be of less importance.
However, major disasters or changes in deltas and small islands could have repercussions over large areas. Given the populations potentially involved, this is more likely to seriously affect the major deltas, such as the Ganges-Brahmaputra, Mekong and Nile. In both the cases of deltas and small islands, a likely scenario could be outmigration when disasters due to sea-level rise reach levels or frequencies considered unacceptable. It is at such thresholds that maximum damage and loss of life could be expected.
It is also noteworthy that deltas, although not specifically fragile, are the result of a long evolution and that they are, in the words of Riebsame et al. (1995), tuned to the current climate. Any departure from the current equilibrium, even one that would a priori be positive (e.g. better rainfall/evaporation ratio), results in a temporary disturbance of the ecology and productivity.
It may be useful to more closely examine the evolution of coastal lowlands during the period from about 10,000 to 6,000 years ago: the sea level rose faster than ever since, by some 125 m in 4000 years. In comparison, the sea level on the coast of the Netherlands rose by 4 m over the last 4000 years.
Speculating on the impacts of that extremely rapid and prolonged sea level rise on human society, we might note that near its beginning, some 9000 years ago, arable agriculture began. Just after the end, some 6,000 to 5,500 years ago, there were suddenly a series of well planned cities, with well ordered agricultural areas surrounding them, in various coastal lowlands or lower river valleys, e.g. in the Euphrates-Tigris, the Nile, the Indus.
Where are the precursors of these cultures, so sophisticated in their spatial planning? Did they originate near the coast much earlier, at lower elevations, with those societies that survived and prospered being those that could best plan their successive evacuations to higher ground, re-establish their physical infrastructure and maintain their social organization? A re-reading of the several "Flood" records in ancient literature worldwide might provide new insights. Also, subaqueous investigation of the shallow parts of the relevant undersea deltas - for example, by seismic means or possibly ground-penetrating radar - might yield new information on that intriguing period. We might learn something of use to present society, which is facing similarly great changes, albeit of a different kind.
World-wide data; graph prepared using data from Tobler et al. 1995 and 1997. The 9 regions covered are Warm humid tropics, Warm seasonally dry tropics, Cool tropics, Arid regions, Subtropics (summer rains), Subtropics (winter rains), Temperate zone (oceanic), Temperate zone (continental) and Boreal zone.
Finally, one observes marked differences in population growth according to agroclimatic zones. Population densities grow fastest in the seasonally dry tropics, while in temperate countries little change takes place. This will lead to the need to produce more food in the former zones, which have quite sever ecological and agricultural limits (Gommes, 1992).
5. The "multiplier" expresses how many times the coast is more dissected than would be estimated from the length measured as straight lines "as the crow flies".
6. The main deltas are the Ganges-Bahmaputra delta in Bangladesh, the Nile in Egypt, and the Mahanadi and Ganges in West Bengal. There are about nine major deltas in the Americas (two flowing into the glacial Ocean and seven in sub-tropical and tropical areas), three in Europe, one in glacial Asia, 10 in tropical Asia, four in Australasia and five in Africa (Jelgersma et al 1993).
To: Cover page | 1. Introduction / setting | 2. Effects of sea-level rise | 3. Lessons from the past | 4. Vulnerability index | 5. Conclusions, references | 6. Appendix table: AOSIS countries
© FAO 1998