A rise in average temperature, increasing rainfall and changes in circulation patterns are accompanied by changes in the frequency and intensity of extremes. It is no surprise, therefore, that the number of weather-related disasters has risen rapidly over the last few decades. Furthermore, from the 1960s to the 1990s economic damage caused by weather-related disasters increased eightfold even after correcting for inflation. Much of this can be explained by the fact that there are more people with greater assets, by the growth of towns and cities in vulnerable areas, and similar factors. But probably some 10 per cent, and maybe up to 50 per cent, of the rise in the cost of weather-related damage is caused by climate change. Precise attribution is very complex as the climatic factor varies by region and by type of event. For example, stronger and more frequent El Niños are accompanied by fewer hurricanes on the east coast of the United States of America, but cause more damage elsewhere in the world.

Given the momentum of climate change, and the projections of global greenhouse gas emissions, the increasing cost of weather-related damage is very likely to continue over the next few decades. Present trends suggest that it will rise by $50 to $100 billion annually over the coming ten years.

Reducing risks
Cutting the growth of global emissions would reduce the rate of climate change, so that people and ecosystems would find it less difficult to adapt, and would suffer less. Perhaps more importantly, it would reduce the risk of destabilizing the global climate. Such destabilization – a relatively rapid change over a matter of decades – could come about as a result of major changes in the patterns of ocean and atmospheric circulation, or in the volume of snow and ice, or through the triggering of major positive feedback from the release of ‘natural’ sources of greenhouse gases.

The major risk to the climate – quite different from the gradual rise in temperature and sea level – is the possibility of fast, flip-flop changes in it. Small changes in the energy balance of the Earth can trigger large changes in atmospheric and ocean circulations and these can have major consequences on temperature, rainfall, wind and storm patterns, and extreme events.

The probability of such changes occurring in the climate’s regime may be low over the next 100 years – and they are difficult to predict – but when they do happen they will have a major impact on life on Earth. Evidence that they have occurred in the past has been found in paleoclimatic records. Some impacts might be reversed when atmospheric concentrations of greenhouse gases decrease again; but others could be largely irreversible, as the climate system has many equilibria.

One such possible fast change could shut down Europe’s natural heat pump, through the stagnation of the Ocean Conveyor Belt – a thermohaline circulation driven by differences in the density of seawater controlled by temperature and salinity. This conveyor belt transports an enormous amount of heat northward, through the Gulf Stream, making the climate of north-western Europe on average 8 degrees centigrade warmer than the mean value at its latitude. The water cools and sinks in the North Atlantic; it then moves south and circulates around Antarctica, before going northward to the Indian, Pacific, and Atlantic Oceans basins. Density differences between seawater determine the ‘strength’ of this circulation: a change in them, as a result of climate change, could lead to a ‘weakening’ or even a stagnation of this ocean current.

The Conveyor Belt circulation pattern is susceptible to perturbations resulting from injections of excess freshwater (from precipitation or melting ice) into the North Atlantic. At present the sinking of saltwater near Greenland ‘pulls’ less saline warm water to the North Atlantic Ocean. But increasing rainfall at higher latitudes would make this surface water less salty, so it would sink less and the strength of the conveyor circulation could decrease.

Unimaginable consequences
Computer models of the world’s climate indicate that an increase of precipitation at higher latitudes, as a result of global warming, could shut down the conveyor over less than a decade within 100 to 300 years from now. Ice core records indicate that, when such cessations have happened in the past, there was a temperature drop of 7 degrees centigrade. One recent study – based on a computer model which for the first time gives a realistic simulation of large-scale ocean currents – published in the magazine Nature, suggests that after an initial rise due to the greenhouse effect, average temperatures would suddenly and rapidly drop by about 6 to 8 degrees. This would disrupt ecosystems and economic systems to an extent beyond imagination, and have dramatic consequences for the people of the northern hemisphere, particularly Europe.

At the other end of the Earth, the West Antarctic icesheet also poses a high impact but low probability risk. Ice shelves are already breaking up at the Antarctic Peninsular as a result of warmer annual and summer temperatures and a lengthening of the season when the ice melts. On 15 October 1998, an iceberg one and a half times the size of the state of Delaware ‘calved’ off from the Ronne Ice Shelf.

A rapid control and reversal of the growth of global emissions would reduce the rate of climate change

Most climate models indicate that there will be modest temperature rises around Antarctica over the next 50 years: over this time period, increased precipitation will probably more than compensate for increased surface melting. But after that, if temperatures continue to rise, Antarctica may start to warm enough for there to be a significant impact on particularly vulnerable parts of the Ice Sheet. Interpretations of present and future changes in Antarctica are very complex, and sometimes contradictory, but it is clear that small changes in the West Antarctic Ice Sheet could cause the seas to rise around the world by several metres, inundating coasts and cities.

The likeliest scenario, if emissions of greenhouse gases continue to increase, is that the Ice Sheet will disappear over the next 500 to 700 years, but there is a very small chance that this could happen in the next 100 years. Sea level would then rise abruptly by between 4 and 6 metres. As temperatures rise, the Greenland Ice Cap would also become vulnerable. This is expected to be irreversible once it begins to melt, and would lead to seas rising several metres further.

Feedback mechanisms
Three other mechanisms, which also combine a small chance of occurrence with far-reaching consequences, depend on positive feedbacks which accentuate global warming.

The cool and humid climate of northern Europe and Asia, (the boreal zone) for example, has helped carbon to accumulate in its soils. There is between 1.2 and 1.5 times as much of the element in these soils as in the atmosphere, and indeed about 40 per cent of all the carbon that is estimated to be stored in the world’s forests is in the boreal zone. Global warming could cause much of this stored carbon to be released, leading to a positive feedback and accelerating the rise in global temperatures.

Similarly the rising temperature of the oceans as a result of global warming could, in another positive feedback, release dissolved carbon dioxide. Instead of absorbing the greenhouse gas, as at present, the seas could cause concentrations of it to increase even further in the atmosphere, accelerating climate change.

Another possible feedback concerns methane clathrate, an ice-like compound in which methane molecules are caged within cavities formed by water molecules. This forms in the oceans in continental slope sediments, and is stable under certain combinations of temperature and depth. However, it could become unstable with warming and release enormous amounts of methane, itself a potent greenhouse gas, leading to a complete destabilization of the present climate. It should be said, however, that this model is somewhat speculative, and is based on a number of assumptions that have yet to be verified.

Destabilization of the global climate is a real risk. The probability of it happening in the short term is low, but it becomes much more likely over the longer term – and it would have extremely far reaching consequences. As climate is a unique, very complex and relatively unknown system, such potential phenomena should play a central role in developing strategies for controlling greenhouse gases.