The 1995 IPCC Second Assessment Report estimated changes in the number of people flooded by storm surges due to a one metre sea-level rise, and the losses in coastal wetlands. The rise in sea level assumed was just above the top end of the range for 2100 suggested by IPCC, and the calculations did not consider the rapid socio-economic changes which are occurring in the coastal zone. To estimate the impacts of sea-level rise on coastal flooding and coastal ecosystems more realistically, we consider the effects of three simultaneously changing factors: global sea-level rise, coastal population, and upgrading flood protection due to increasing national wealth. It should be noted that storm surge climatology is assumed to be constant. If the number of coastal storms increased, the impacts of flooding would be exacerbated and vice versa.

Coastal Flooding
To estimate the number of people affected by flooding for a given sea-level rise scenario, a model has been developed which includes a database at a nation scale of present coastal elevation, subsidence, storm surge characteristics, flood protection standard and coastal population density. (The standard of flood protection is estimated using GNP per capita as an ‘ability to pay’ parameter as no suitable global databases are available). Unlike earlier work, the decline in protection as sea level rises is evaluated. It is further assumed that all coastal populations are presently protected to at least a 1-in-10 year standard. Hence, the area subject to flooding for different global rises of sea level can be calculated. Model results have been validated for the present situation and a one-metre rise in sea level against six national assessments (Egypt, Germany, Guyana, Netherlands, Poland and Vietnam). Global sea-level rise was estimated from the thermal expansion calculated by the Hadley Centre climate model and ice melt contributions quoted in IPCC 1995 (IS92a scenario): the total rise by the 2080s is predicted to be 44 cm. It is assumed that the coastal population will grow at the current trend of twice the national rate.

The results are described using the number of people at risk, which is defined as the average number of people flooded per year by storm surge. Presently the number of people at risk is approximately 10 million, rising to about 30 million in the 2080s under the reference scenario of constant (1990s) flood protection and no sea-level rise. Given the sea-level rise scenario, the number of people at risk increases dramatically being about 700% above the reference scenario by the 2080s (diagram, top left). However, even without sea-level rise, present experience suggests that flood protection will be upgraded in flood-prone countries as national wealth rises, and this will reduce the number of people at risk. As a first estimate of these change, it is assumed that flood protection is upgraded in line with the projected increase in GNP per capita (but with no additional allowance for sea-level rise). The number of people at risk will still increase by about 250% relative to the reference case by the 2080s (see top figure). Many of these people will experience flooding more frequently than once per year, and it seems likely that they will need to respond in some way (e.g. migration, upgraded sea defences, etc.). Collectively, these results show more dramatic impacts than described by the 1995 IPCC Second Assessment Report. They suggest that the sea-level rise scenario will cause significant flooding impacts in the coastal zone, particularly in deltaic countries, unless specific adaptation measures for sea-level rise are taken.

Most of the people at risk in the 2080s are concentrated in a few regions, particularly the southern Mediterranean, Africa, southern Asia and South East Asia (see the map at bottom opposite). In addition, relatively high numbers of people at risk are found in five nations which comprise low-lying coral atolls in the Indian and Pacific Oceans (more than 30% of national population in the 2080s assuming 1990s protection).

Coastal Wetlands
Coastal wetlands, comprising saltmarshes, mangroves and intertidal areas, are sensitive to sea-level rise as their location is intimately linked to sea level. However, they are not passive elements of the landscape and, as sea level rises, so the surface of any coastal wetland rises due to sediment and organic matter input. If this rise keeps pace with sea level the coastal wetland will grow upwards in place, but if it does not, the wetland steadily sinks relative to sea level. Intertidal areas will be steadily submerged. Vegetated wetland systems will be submerged during a tidal cycle for progressively longer periods and may die due to waterlogging, causing a change to bare intertidal areas, or even open water. Therefore, coastal wetlands show a dynamic and non-linear response to sea-level rise. Coastal wetlands with a small tidal range are more vulnerable than those with a large tidal range. Direct losses of coastal wetland due to sea-level rise can be offset by inland wetland migration (upland conversion to wetland as sea level rises). In areas without low-lying coastal upland, or in areas which are protected to stop coastal flooding, wetland migration cannot occur.

The direct wetland response to sea-level rise is modelled by selecting two critical values of sea-level rise, scaled by tidal range; the lower value distinguishes no wetland loss from wetland loss; while the upper value distinguishes partial loss from near-total loss. Loss is modelled linearly between the two threshold values. The potential for wetland migration on to adjacent low-lying upland is evaluated, based on coastal morphology and coastal population density. In addition to the effects of sea-level rise, direct human reclamation is likely to cause large global reductions in coastal wetlands. Based on current trends, 60% of the present wetland stock would be lost by the 2080s without consideration of sea-level rise. It is likely that the loss rate of coastal wetlands will decline with time due to both an increasing rarity, and rising living standards which given the environment a higher ‘value’. Therefore, a reference scenario of losses of 1% a year in the 1990s, declining uniformly to a constant 0.4% a year in the 2020s, was assumed. This gives a loss of 37% of the global wetland stock by the 2080s without sea-level rise.

The middle figure opposite shows the decline in the global stock of wetlands as a function of time. Due to uncertainties concerning model parameters, a low-loss and a high-loss estimate are shown defining a large range of uncertainty. The losses due to sea-level rise are negligible before the 2020s. By the 2080s, the sea-level rise scenario causes the loss of up to 25% of the world’s coastal wetlands. When combined with the direct loss scenario, about 40-50% of the world's coastal wetlands could be lost by the 2080s. Losses due to sea-level rise vary substantially from region to region. As shown in the map on the opposite page, coastal wetlands are most threatened on the Atlantic coast of North and Central America and around the Mediterranean and the Baltic. This is due to the low tidal range in these areas combined with the limited potential for wetland migration.

Loss of wetlands will impact many sectors including food production (loss of key nursery areas for fisheries), flood and storm protection (storm surges will penetrate further inland), waste treatment and nutrient recycling functions, and as habitat for wildlife. These results stress the need for future effective wetland conservation worldwide.

The commitment to sea-level rise
An important aspect of rising sea level is its long timescale. Sea level will rise as meltwater from land ice runs into the ocean and as the ocean waters expand. Even if climate change could be halted, the surface warming already incurred will progressively penetrate deeper and deeper into the ocean, causing sea level to rise still further. The figure above shows how the thermal expansion component of sea-level rise in the Hadley Centre climate model will increase fivefold over the 500 years after all climate change is stopped, and will continue to increase for centuries further. Therefore, sea level will continue to rise after the 2080s, even if further climate change is halted. While impacts have not been evaluated in detail, it seems likely that unavoidably greater losses of wetlands will occur beyond the 2080s.

Conclusions

• Without specific adaptation, sea-level rise will significantly increase the flood risk to coastal populations.

• In absolute numbers, the southern Mediterranean, Africa, southern Asia and South East Asia are most vulnerable to the impacts of flooding.

• Coastal wetlands are expected to decline due to sea-level rise, with the largest losses around the Mediterranean and Baltic and on the Atlantic coast of Central and North America.

Climate change science

• The 1997/98 El Niño was the most extreme on record. The global mean surface temperature in 1998 is likely to exceed that in 1997 and be the highest since global instrumental records began.

• Predictions of climate change have been made using an improved climate model without corrections. IPCC ‘business-as-usual’ increases in greenhouse gas emissions result in a further warming of about 3 ºC over the next 100 years.

• Comparisons of model simulations and observations indicate that human-made greenhouse gases have contributed substantially to global warming over the past 50 years. Initial results also indicate that climate models can simulate reasonably well the climate change of the past 140 years, and this gives us confidence in predictions of the future.

• The new climate model has a better representation of ocean currents. Increases in greenhouse gases result in a slowing down of the North Atlantic ocean circulation, but even with this, Europe still warms.

Impacts of climate change on natural vegetation

• Under the latest climate scenario for the 2050s, tropical forests will die back in many areas of northern Brazil. In other areas of the world, tropical grasslands will be transformed into desert or temperate grassland.

• Vegetation will absorb CO₂ at the rate of some 2-3 GtC per year in the first half of the next century; this compares to current human-made emissions of about 7 GtC per year. After 2050, and as a result of vegetation die back, this will become a source of about 2 GtC per year, thereby enhancing CO₂ build-up in the atmosphere. This enhancement is not yet included in climate predictions.

Impacts of climate change on water resources

• Water resource stresses in many of the poorest countries, already expected to increase, will be exacerbated by climate change. Due to climate change alone, some 66 million extra people will live in countries with water stress, and some 170 million people will live in countries which are extremely stressed.

Impacts of climate change on food supply

• Due to changes in climate and CO₂ crop yields are expected to increase in high- and mid-latitude countries but decrease in lower latitudes. Although globally the food system will accommodate regional variations in yields, some regions, particularly in the Tropics, will experience marked reductions in yield, lower production and higher risk of hunger. Africa will be worst affected, with 18% more people at risk of hunger due to climate change alone.

Impacts on coastal communities

• Global mean sea-level rise by the 2050s is predicted to be 21 cm. If coastal protection evolves only according to Gross Domestic Product (GDP), as in the past, then over 20 million extra people each year will be at risk of flooding due to sea-level rise.

• The coasts of the southern Mediterranean, West and East Africa, east Asia and, most particularly, south and South East Asia are most vulnerable in absolute terms. The islands of the Caribbean, and the small islands of the Indian Ocean and Pacific Ocean, are also vulnerable in terms of large relative increases in flood risk.

Impacts of climate change on human health: malaria

• While growth in population will itself increase the number of people at risk of malaria, climate change will increase the proportion of the world population at risk, particularly in areas where currently the disease is not endemic.