Storm brewing, southern France

A greenhouse effect has always existed, keeping the Earth warmer than it would be without an atmosphere. What is popularly referred to today as the 'greenhouse effect' is really the anthropogenically enhanced greenhouse effect by which an extra warming of the surface and the lower atmosphere is produced, leading to disturbances in the geosphere/biosphere system and, notably, an increase in the mean global surface temperature and in the mean sea level.

The natural presence of 'radiatively-active' gases in the atmosphere is essential for life: they trap heat in the lower atmosphere, thus creating ­ like a greenhouse ­ an environment which is far warmer (by about 33°C) inside than out. By increasing the concentration of greenhouse gases, however, additional infra-red radiation, which otherwise would have been lost to space, is absorbed in the lower atmosphere and the Earth's radiation balance is upset. This energy is re-emitted in all directions, a large portion being sent back to the Earth's surface or elsewhere in the troposphere. This yields a radiative imbalance which can be restored only through a warming of the troposphere. On the other hand, the enhanced greenhouse effect will also cool the upper layers of the atmosphere (ie, the stratosphere, and above between 25 and 70 km).

Greenhouse gases in the atmosphere have increased since pre-industrial times by an amount that is radiatively equivalent to about a 50 per cent increase in carbon dioxide (CO₂), although CO₂ itself has risen by about 25 per cent; other gases have made up the rest. The global emissions of most greenhouse gases are expected to rise in the next decade: CO₂ at 0.5 per cent/year, methane (CH₄) at 0.9 per cent/year, and nitrous oxide (N₂O) at about 0.3 per cent/year. In contrast, CFC emissions are expected to decrease to near zero around the year 2000 following internationally agreed phase-out measures. As a result of these trace gas trends, an effective doubling of greenhouse gas concentrations (as commonly measured as 'CO₂-equivalents') is expected around 2030.

There has been much scientific activity aimed at identifying clues of the enhanced greenhouse effect by searching for trends in climatic and hydrological records (mainly temperature, precipitation, glaciers, runoff, freezing/melting date and sea level). Conclusions have so far been mainly tentative because definite signals from an enhanced greenhouse effect are weak and/or smaller than natural fluctuations, most of the homogeneous and comparable records, if any, are too short, and regional differences have added to the confusion (see Chapter 4). Our partial understanding of multimedia mechanisms involved has also been a limitation to making a safe forecast of future trends, although this has fostered large international research efforts such as the World Climate Research Programme (WCRP, of the World Meteorological Organisation (WMO), the International Council of Scientific Unions (ICSU) and the Intergovernmental Oceanographic Commission (IOC) of UNESCO) and the International Geosphere-Biosphere Programme (IGBP of ICSU), or international expert assessments such as under the Intergovernmental Panel on Climate Change (IPCC of WMO and UNEP).

At first glance, uncertainties about the distribution and timing of consequences of climate change may make some people feel that the cost of rapid implementation of actions intended to curb the man-made emissions responsible might be prohibitive. However, regardless of what happens to climate, application of relevant measures would bring its own benefits, such as longer availability of fossil fuels and other natural resources, as well as more rational and efficient energy use. Additionally, consequences such as stronger and more frequent storms, higher seas and less water in rivers are so far-reaching that taking this 'precautionary' approach should be rewarding. This approach is also known as the 'no regrets' policy, because of the argument that preventing climate change would be far better than attempting to cure it, despite the uncertainties of how fast and how soon it might be happening.

Although Europe's land mass covers less than 10 per cent of the world's total area and its population is some 15 per cent of the world population, European countries contribute a significant amount to the global emissions of greenhouse gases and other substances which bring about large-scale changes in the atmosphere (see Chapters 4 and 14). Conversely, climate change may greatly affect Europe's 680 million people and its natural environment.

Human exploitation of the world's natural resources, primarily fossil fuels ­ coal, oil and natural gas ­ but also biomass resources through agricultural and forestry practices, result in the release each year of over 20 000 million tonnes of carbon dioxide (CO₂) into the atmosphere. In addition, other atmospheric gases, notably methane (CH₄), nitrous oxide (N₂O) and chlorofluorocarbons (CFCs), are being released as a result of human activities. All these gases together with ozone (O₃) and water vapour are referred to as 'greenhouse gases', that is, they are relatively transparent to incoming short-wave radiation from the sun, but absorb long-wave infra-red radiation emitted from the Earth (Figure 27.1), and consequently entrap heat in the atmosphere. Ozone in the lower atmosphere (troposphere) is not emitted directly but is formed from anthropogenic emissions of nitrogen oxides (NO_x), volatile organic compounds (VOCs), methane and carbon monoxide. Particles (eg, sulphate) emitted into the atmosphere by anthropogenic and natural sources (mainly volcanoes) can also affect climate because they can reflect and absorb radiation.

Various physical and dynamic processes combine to redistribute heat horizontally and vertically in the atmosphere and the oceans. At the same time, they trigger various feedback effects that can either enhance or dampen (reduce) the initial trends. Most of these feedback processes are connected to the formation of clouds as well as to biospheric and oceanic responses.

Soils are an important reservoir for carbon; an estimate of the global soil carbon pool is 1500 gigatonnes (billion metric tonnes), or twice the atmospheric pool, excluding carbon in organic soils. Globally, the amount of carbon in decaying plant litter and soil organic matter may exceed the amount of carbon in living biomass by a factor of two or three. Soils also contribute to the natural fluxes of greenhouse gases; the contribution of the whole terrestrial biota (soil, vegetation and fauna) is 30 per cent for CO₂, 70 per cent for CH₄ and 90 per cent for N₂O.

Recent estimates suggest that 2 gigatonnes of carbon are absorbed by the oceans each year to replace the carbon taken by phytoplankton from sea water and subsequently deposited on the sea bed when the phytoplankton die.

The lifetime of greenhouse gases in the atmosphere is determined by their sources and sinks in the oceans, atmosphere and biosphere. Carbon dioxide, CFCs and nitrous oxide are removed only slowly from the atmosphere (see Table 27.1) and hence, following a change in emissions, their atmospheric concentrations take decades or centuries to adjust fully. Even if all human-made emissions of CO₂ had been halted in the year 1993, about half of the increase in carbon dioxide concentration caused by human activities would still be observable in the year 2100.

The concept of relative global warming potentials (GWPs) has been developed as an index of the relative radiative effect (and, hence, potential climate effect) of equal emissions of each of the greenhouse gases, to take into account the differing time that they remain in the atmosphere and their different absorption properties. The GWP defines the time-integrated warming effect due to an instantaneous release of unit mass (1 kg) of a given gas in today's atmosphere, relative to that of carbon dioxide. The relative contributions will change over time, and a period of 20 years will indicate the response to emission change in the short term (see Table 27.1). Anthropogenic CO₂ emissions are currently responsible for more than half of the enhanced greenhouse effect (Figure 27.2). However, the GWP concept does not allow for the comparison of the relative effect of different greenhouse gases in terms of their contribution to climate change per se, such as may be measured in terms of the increase in the global mean surface temperature or sea-level rise. This comparison can only be made with the use of complete atmospheric models, and simple comparative indices are now being developed.

During the past century, the concentration of the most important greenhouse gases, and notably that of carbon dioxide, has been increasing steadily, showing that anthropogenic emissions have exceeded the natural capacity for their removal either through absorption at the Earth's surface or through chemical reactions in the atmosphere. Figure 27.3 illustrates the global CO₂, N₂O, CH₄ and CFC-11 concentrations since 1750.

The use of fossil fuels for energy production and transport is the most important source of global CO₂ emissions (Table 27.2). Emissions due to coal are about twice those of natural gas for the same amount of energy (oil-related emissions are about one-and-a-half times as much). Present conversion of tropical forests to agriculture and pasture releases up to 15 per cent of global emissions (approximately 7000 million tonnes of carbon in 1991).

Currently, European forests are a net sink for CO₂ as a result of afforestation and probably because of a CO₂-fertilisation and nitrogen-fertilisation effect. In most cases, the conversion of native forests to plantations will have resulted in a net loss of carbon to the atmosphere, including the carbon stored in timber products (Cannell et al, 1992).

Ordinary agricultural practices also play a role in the problem of climate change. The yearly burning of biomass releases an amount of carbon, in the forms of carbon dioxide, carbon monoxide and methane, which is smaller than, but of the same order of magnitude as, the emissions from the burning of fossil fuels. The carbon released from biomass burning, however, like that from the decay of fallen leaves in temperate forests, does not correspond to a net release in the long term, since the same amount of carbon is removed from the atmosphere in the following growing season. Indeed, the oxidation of carbon at the higher temperature of burning tends to generate a higher percentage of carbon dioxide, which produces a smaller enhancement of the greenhouse effect, molecule by molecule. The burning of biomass, on the other hand, does affect the chemistry of the troposphere, notably by increasing the ozone concentration, which may also have local pollution effects.

Scientific assessments of global climate change have generally considered only long-lived greenhouse gases merged into one equivalent surrogate (CO₂), which is assumed to double homogeneously by the middle of the next century. The IPCC reports (IPCC, 1990 and 1992) mentioned the greenhouse property of ozone, but made no quantitative assessment of its warming potential. In effect, because ozone does not last long in the atmosphere, this assessment is more difficult since both temporal and spatial distribution should be taken into account. A recent estimate (Marenco et al, 1994) computes a relative radiative forcing for ozone at least 1200 times higher than for CO₂, that is, much higher than the other greenhouse gases except CFCs. With its current concentrations, ozone would thus contribute 22 per cent of the global warming in the northern hemisphere and 13 per cent in the southern hemisphere.

The increase of methane concentrations is correlated largely with increasing population, about 60 per cent of global emissions being associated with human activities. Methane from fossil origins is released by the exploitation of coal, oil (if it is not used or flared) and natural gases, and from distribution systems. As a product of the anaerobic digestion of organic material, it is also released from landfills, wetlands and rice paddies and by fermentation in the rumen of ruminants. Because of the potency of methane in the atmosphere and its relatively short lifetime, stabilising CH₄ concentrations will have a substantial impact on reducing potential warming.

Nitrous oxide emissions are produced mostly by denitrification processes in oxygen-free environments with a high nitrate load, such as soils and sediments in polluted waterbodies. N₂O is also released in limited quantities by the use of fossil fuels.

Carbon monoxide plays significant roles in controlling the chemistry of ozone production and hydroxyl radical OH destruction in the lower atmosphere. It directly affects the oxidising capacity of the lower atmosphere and thus influences the concentrations of other important trace gases such as methane, methylchloroform and the HCFCs.

Global anthropogenic emissions of sulphur gases have increased by about a factor of three during the past century, leading to increased sulphate aerosol concentrations, mainly in the northern hemisphere. Sulphate aerosols can affect the climate directly, by increasing the reflection of incoming solar radiation back to space in cloud-free air, and, indirectly, by providing additional cloud condensation nuclei. Model calculations show that the increase in sulphate aerosol concentration reaches a factor of 100 over Northern Europe in winter.

The primary consequence of increased radiation absorption is climatic change or, in terms of the most common climate indicators, rapid changes of surface temperature ­ about 0.3°C/decade (IPCC, 1992) ­ and precipitation changes. The consequent expected sea-level rise (3 to 10 cm/decade) and changes in hydrological and vegetation patterns may have serious effects on society, leading to high risks and to substantial costs. Current models predict average global changes, but the impacts of climate change will be most keenly felt on a regional and local scale where uncertainties are larger (see Table 27.3).

To evaluate the consequences of climate change, it is common to take as a reference case the doubling of CO₂ (or an equivalent amount of total greenhouse gases) in the atmosphere relative to its pre-industrial level (before the years 1750­1800). The IPCC business-as-usual scenario (which is considered a high emission scenario) suggests that this will occur by 2025. Under the IPCC low emission scenario, doubling will take place around 2060.

General circulation model estimates of the increase in the global annual average temperature at the surface of the Earth vary from 1.5 to 4.5°C, given an increase in greenhouse gases equivalent to a doubling of the pre-industrial CO₂ concentration. The current 'best' estimate is 2.5°C (IPCC, 1990). The average global increase in precipitation and evaporation is estimated at 3 to 15 per cent. A recent reassessment of the IPCC90 climatic trends, which also takes into account the effects of CO₂ fertilisation, feedback from stratospheric ozone depletion and the counteracting radiative effects of sulphate aerosols, yields new projections for radiative forcing of climate and for changes in global-mean temperature and precipitation (Wigley and Raper, 1992). Changes in temperature and sea level are predicted to be less severe than those estimated previously, but are still far beyond (four to five times) the limits of natural variability.

The range of possible rise in temperature is determined largely by the uncertainty about the feedback processes. While the timing and extent of warming on a global basis is uncertain, still more indefinite is the change of climate on the regional level in Europe and elsewhere. The spatial resolution of models being still too coarse, regional estimates are less accurate than global ones. Models, however, consistently predict that temperature increases will be greater in the high latitudes than low latitudes. Hence the extent of temperature increases in Northern Europe are expected to be larger on average than those in the Mediterranean regions. Using the new technique of nesting computer models, improved estimates of possible temperature changes in Europe are becoming available (see Map 27.1).

The IPCC collected best model guesses as to the change in climate for Southern Europe from pre-industrial times to 2030, under the assumption that average global warming would be 1.8°C by that year. Temperature was estimated to be 2°C warmer in winter and 2 to 3°C warmer in summer, precipitation 0 to 10 per cent greater in winter and 5 to 15 per cent lower in summer, with a ­5 to +5 per cent change in soil moisture in winter and ­15 to ­25 per cent change in summer. The ranges in these values indicate the disagreement between the three different general circulation models used for these calculations. As can be seen (Map 27.1), the models are relatively consistent, although in some cases they do not even agree on the direction of change.

Higher temperatures will yield more evaporation, resulting in more precipitation. This will intensify the hydrological cycles, with unknown consequences. Sudden regional or global climate changes following changes in the location and/or intensity of air and water flows cannot therefore be excluded.

Many studies on specific climate change impacts have not used a common framework, and some of their underlying assumptions may differ. As yet, there has been no comprehensive, systematic study of climate impacts in Europe. However, preliminary results from a still unpublished study of the European Commission (performed by the Climate Research Unit, East Anglia, UK; Environmental Resources Management Limited, London, UK; and the National Institute of Public Health and Environmental Protection (RIVM), Bilthoven, The Netherlands) indicate that the possible changes in the EU due to global warming would be increased agricultural yields and tourism, and reduced heating requirements. However, the study also points out that the monetary costs of protecting coastlines against sea-level rise are likely to far outweigh benefits.

A rise in the global mean sea level will be one important impact of global warming. It will result from:

1. the thermal expansion of sea water;
2. melting of mountain glaciers;
3. melting of inland ice caps, as in Greenland; and
4. changes in the Antarctic ice sheet.

According to IPCC best estimate scenarios, the global-mean sea level would be about 22 cm higher than today by 2050 and around 50 cm higher by 2100 (the low and high scenario estimates give 15 and 90 cm respectively for 2100). A rise of 10 to 20 cm over the last 100 years has been observed, due mainly to oceanic thermal expansion and retreating mountain glaciers.

The consequent environmental effects are many and varied: permanent inundation of low-lying land; increased frequency of temporary flooding from high tides or storm surges; changes in rates of beach, dune or cliff erosion; salinisation of estuaries, groundwater and surface water supplies, wetland ecosystems or agricultural soils; and effects on river hydrology, including inland flooding, through changes in river gradients. The most threatened areas in Europe, because of their location and elevation, are shown in Map 27.2.

Among the greatest potential impacts of climatic change will be the effects on the hydrological cycle. Increase in incidence of extremes, such as floods and droughts, would cause increased frequency and severity of disasters, while changes in precipitation, evapotranspiration and soil moisture would strongly alter agricultural practice and water management systems, and lead to severe land degradation. Forecasts of future changes of water resources attributable to human-induced climate changes are still uncertain because of the unreliability of regional precipitation predictions. However, they do indicate that such changes would be fairly significant, even in the case of slight climate modification, and that not only river runoff but also water demand might be affected. For instance, a temperature rise of 1 to 2°C and a 10 per cent drop in rainfall would reduce river runoff in arid regions by 4 to 70 per cent (Shiklomanov, 1991).

Various case studies point out possible modifications in runoff and other processes that might occur if there were a doubling of CO₂. For example, a study of three catchments in Belgium (Bultot, 1988) indicated that winter floods may become more frequent in catchments that have low infiltration rates and unsteady base flows; however, in basins having high infiltration rates and more steady base flow, there might be a beneficial increase in base flow throughout the year. In a study of seven Norwegian stream basins, climate changes brought about by a doubling of CO₂ are estimated to increase runoff in mountainous catchments, while decreasing it in lowland basins because of greater evapotranspiration. The seasonal pattern of runoff will change markedly, particularly in catchments of intermediate elevation. Winter runoff will increase significantly, while summer runoff declines. Flooding will occur more frequently in autumn and winter, and the overall amount of flood damage is expected to increase. The net effect of changes in runoff patterns may also lead to a small increase in Norwegian hydropower production.

Preliminary model results have shown an increase in runoff (up to a doubling) of rivers discharging into the Arctic Ocean. Since freshwater runoff into the Arctic Ocean is a significant fraction of its total water mass, any substantial modification of the input of water as envisioned in model scenarios may change important thermal and salinity characteristics. This in turn could significantly modify currents, ice cover and atmospheric circulation. Changes in river flow also modify the delivery of dissolved constituents and particulates to the Arctic Ocean. Increases in nutrient inputs could increase net primary production and thereby sequester excess atmospheric CO₂.

Alteration of the hydrological cycle, with changes in runoff and moisture availability, will influence patterns of sedimentation and erosion and the recycling of organic matter and nutrients. These will in turn influence plant productivity, competition between species and biodiversity. The linkages are sometimes more subtle, with, for example, changed temperature and precipitation conditions combining to favour the outbreak of plant diseases or insect plagues. These complex linkages between different factors, together with uncertainties in predicted water availability, make it difficult to predict changes in natural ecosystems towards new equilibria. Moreover, changes may take place so quickly that plants have no time to spread to new habitats by natural dispersion mechanisms and the actual vegetation response could therefore differ markedly from equilibrium predictions.

Climate changes will influence ecosystems and agriculture since temperature and evapotranspiration/ precipitation determine the major biogeographical zones (such as northern boreal, temperate, Mediterranean). Large regional differences may arise due to small changes in one of the critical factors, such as temperature, precipitation, evaporation and soil structure, yielding changes in vegetation structure and the range of species of a given zone. The potential impacts of climate change on agriculture may have dramatic consequences since Europe, as a main exporter of agricultural products, contributes significantly to the nutritional balance of non-European countries. This possibility for a major disturbance in agricultural trades should be seen also in the context of a rapidly increasing world population.

A study of the potential range of maize-growing areas in Europe found that a 1°C increase in mean annual temperature in Europe would translate into a northward shift of approximately 200 to 350 km in Western Europe, and 250 to 400 km in Eastern Europe. New areas of potential maize production would be opened up in southern England, The Netherlands, Belgium, northern Germany and northern Poland. A mean annual increase of 4°C would move the boundary up into northern Russia and central Fennoscandia. However, this study did not take into account possible shifts in water availability, nor possible changes in the range of agricultural pests and diseases, nor other factors that might limit agropotential. Indeed, the IPCC estimated under their business-as-usual scenarios that Mediterranean countries which already depend heavily on irrigation would have 15 to 25 per cent less soil moisture in summer.

Although extensions of agricultural areas northward might prove to be beneficial, the same pressure on natural ecosystems to extend or shift their range may endanger these ecosystems. This is because of human or natural barriers that may prevent the migration of ecosystems as their climatic zones shift. The rate of change is also important because, in less than a hundred years, Europe's bioclimatic zones may shift hundreds of kilometres in latitude (or hundreds of metres in altitude in high Alpine regions), and it is generally accepted that this will place considerable stress on terrestrial ecosystems. In some cases it can be expected that flora and fauna will be unable to migrate fast enough to survive.

Terrestrial ecosystems are thought to play an important role in determining regional and global climate; two examples of this are in Amazonia, where destruction of the tropical rainforest leads to warmer and drier conditions, and the Siberian forests, which represent 40 000 million tonnes of stored carbon (ie, an amount equivalent to half of the forests of Amazonia). Boreal forest ecosystems may also affect climate: as temperatures rise, the amount of continental and oceanic snow and ice is reduced, so the land and ocean surfaces absorb greater amounts of solar radiation, reinforcing the warming in a 'snow/ice/albedo' feedback which results in large climate sensitivity to radiative forcings. This sensitivity is moderated, however, by the presence of trees in northern latitudes, which mask the high reflectance of snow, leading to warmer winter temperatures than if trees were not present. Results from the National Centre for Atmospheric Research (NCAR) global climate model show that the boreal forest warms both winter and summer air temperatures, relative to simulations in which the forest is replaced with bare ground or tundra vegetation (Bonan et al, 1992). This suggests that future redistributions of boreal forest and tundra vegetation (for example due to extensive logging in Russia/Siberia, or the influence of global warming) could initiate important climate feedbacks, which could also extend to lower latitudes. The position of the tree-line distinguishing boreal forest from tundra vegetation has altered in response to past climate changes and is likely to change with the warmer climate caused by increased atmospheric CO₂ concentrations. Results from the NCAR coupled-model raise the possibility of considerable climate changes caused merely by redistribution of boreal forest and tundra ecosystems. The decrease in snow-covered land surface albedo caused by northward migration of boreal forest into tundra zones in response to climate warming may produce further warming.

Plants are the essential basis of all terrestrial life. Any significant variation in the productivity and composition of plant life would initiate a cascade of changes affecting animal life. At first glance, elevated CO₂ levels might seem an agricultural blessing by enhancing plant growth. This 'CO₂ fertilisation effect', as it is called, is expected to be particularly pronounced if plants have plentiful supplies of nutrients, light and water. The CO₂ fertilisation effect also promises to provide a buffer for concerns about global warming by drawing more CO₂ from the atmosphere. However, recent studies (Bazzaz and Fajer, 1992) suggest that such assumptions about the benefits of a world replete with CO₂ may be overstated. An isolated case of a plant's positive response to increased CO₂ levels does not necessarily translate into increased growth for entire plant communities. Farming will also be adversely affected by the differential response of plants. Important crops, such as maize and sugarcane, may experience yield reductions because of the increased growth of weeds.

Even the notion that plants will serve as sinks to absorb ever-mounting levels of CO₂ is questionable. Increased competition between plants will also diminish the CO₂ enhancement of natural ecosystems such as meadows and forests, but also artificial ecosystems such as farms. Agricultural yields will improve in a CO₂-rich future only at the cost of large quantities of fertilisers, pesticides and water from irrigation.

Other organisms which depend on threatened plant species for food, shelter or mating sites may also become endangered. A strong reduction of species diversity would, in turn, undermine the integrity of natural ecosystems. Because individual plant and animal species supply a wealth of essential industrial, agricultural and medicinal products, the loss of diversity will have pervasive environmental and economic consequences. Changes in the nutritional quality of plant leaves could lower herbivore and predator populations within their habitats. If insect herbivores suffer population reductions in a world abundant with CO₂, many predators will have less prey. Some predatory insects, for example, feast on other insect pests that damage certain crops. Plant development and flowering times may be also altered unpredictably by elevated CO₂, thus disrupting pollination.

Shifts in climate and vegetation could also significantly affect future animal distribution, abundance and survival. Rapid rates of change, especially in combination with the existence of artificial urban and agricultural barriers, may affect the ability of many species to relocate to areas which are climatically and ecologically more favourable. Endangered or rare species would be particularly vulnerable to rapid change, especially if their distributions are spatially restricted and their niche-width is narrow. Rapid climatic change therefore becomes a threat for current biodiversity (see also Chapter 29).

The resulting change in atmospheric composition will affect soil biological processes and provoke, for instance, an increase in biomass and changes in composition of vegetation and organic matter. These biological transformations will have an unforeseeable impact on soils. For instance, there is some concern about anticipated effects on soil carbon storage in tundra boreal peatland. On the other hand, human interference in the land-cover brings about changes in the rainfall regime, in the hydrology of large river systems and in the Earth's albedo, all of which play a major role in the surface energy balance.

In the long term (over 50 to 100 years) a substantial increase in temperature and changes in rainfall patterns (perhaps slightly wetter winters) in Europe could modify landuse. Whole ecosystems may shift northwards, but this consequence of climate change will undoubtedly be modified by landuse policy (Cannell et al, 1992). Landuse changes can have such important impacts that soil carbon decreases in the top first metre by 40 to 60 per cent when forest or grassland is converted to cropland, and by 25 to 35 per cent when forest is converted to grassland. Overall, warmer temperatures are thought to result in more arid conditions. Increased aridity in Southern Europe could lead to a chain of events in the soil, beginning with the breakdown in supply of organic matter, continuing with salt accumulation and the formation of surface crusts, and ultimately ending in severe land degradation.

Loss of organic matter may result from removal of crops without the return of any material to compensate. It may also result from alteration of soil drainage, or tillage that accelerates oxidation of organic matter. Peaty soils tend to suffer oxidation of organic matter if drained, and can shrink alarmingly as a consequence. In drier climates, the loss of soil organic matter generally leads to a reduction in retention of soil moisture and a decline in vegetation cover, crops or natural plants, which in turn leads to increased erosion. A positive feedback can thus arise, as organic matter/moisture in soil falls, plant cover declines and thus the renewal of organic matter in soil is further reduced. But for extrapolation of data and making predictions, the underlying processes, which are largely microbial, need to be understood.

Protecting the planet in the long term and maintenance of its ecological equilibrium will require joint efforts to reduce greenhouse gas emissions to a level corresponding to the natural absorbing power of the planet. Emissions should be reduced at such a rate that ecosystems can adapt naturally to climatological changes, and so that food production is not threatened. This requires the most rapid possible stabilisation of the concentration of greenhouse gases in the atmosphere.

One way to quantify the above premises is to determine the acceptable emissions on the basis of a maximum acceptable rate of change of temperature or sea level. Proposals have been made to derive such a limit from the rate of change in the past. The acceptable climatological changes can also be derived from the maximum rate at which ecosystems can adapt to changes in temperature and precipitation patterns. It is generally expected that, assuming a natural limit of the migration rate of terrestrial ecosystems of 50 km per century, a doubling of the CO₂ concentration before 2090 will lead to unstable ecosystems, particularly boreal forests (Map 27.3). This map shows the areas which do not have the same vegetation type across 50 to 80 km and which will be threatened if the CO₂ concentration doubles before 2090 and as climate zones shift by more than 50 km per century. Given these considerations, a temperature rise standard for sustainability of less than 0.1°C per decade has been proposed (Krause et al, 1990). Similarly, a provisional limit of a 2 cm rise in sea level per decade has been recommended to prevent damage to coastal zones, wetlands and coral reefs caused by too rapid climatological changes.

As discussed in previous chapters, the biosphere cannot operate outside certain limits. To remain within these, as yet unknown, limits, absolute limits on the acceptable rise in temperature or sea level could be set. Analyses of past temperature and rises in the sea level show that an absolute temperature change of 2.0°C relative to the pre-industrial era should be considered as a major risk. Considerable changes in ecosystems and sensitive coastal areas, as well as unexpected sudden changes in the climatic system, cannot be ruled out given a temperature change of more than 2°C.

There is a need to improve knowledge of the mechanisms and quantitative components of the budget of relevant compounds if the uncertainty of future projections for the climate and its related components is to be decreased. For example, there are certain greenhouse gases, such as CFC-14 and CFC-116, which have not been included in the IPCC assessment (Abrahamson, 1992). These two gases last for more than 10 000 years in the environment, and have large global warming potentials (being exceeded only by CFC-11, CFC-12, CFC-113 and HCFC-22). Both CFC-14 and CFC-116 are issued from aluminium smelting, but cement is the only primary commodity, other than fossil fuels, specifically included in conventional tabulations on greenhouse gases sources. However, it appears that, on a global basis, aluminium production is an even greater greenhouse contributor than cement production.

Despite the different assumptions behind the rapidly diverging scenarios for the rates of CO₂ emission, there are only very small differences in projected global warming rates over the next few decades. This indicates the commitment to global warming over this time created by past emissions and by current economic development trends. The projected warming to about 2050 is relatively insensitive to changes in global economic activity, or policy initiatives, which may be undertaken between now and 2050. The global warming under the different scenarios begins to diverge substantially by the second half of the next century, so that by 2100 there is a 100 per cent difference between the slowest (1.5°C) and most rapid (3°C) warming projection. This difference represents a significant opportunity for reducing the rate of global warming by the end of the next century, if certain economic and policy choices can be made over the next few years.

To adhere to the goal for sustainability of a maximum temperature rise of not more than 0.1°C per decade, it will be necessary to stabilise the concentration of greenhouse gases in the atmosphere at the lowest possible level, in the shortest possible time. According to the IPCC report the worldwide emissions of greenhouse gases such as carbon dioxide, nitrous oxide (N₂O) and CFCs should be reduced immediately by at least 60 per cent to stabilise the concentrations of these gases at the present level. Methane emissions should be reduced by 15 to 20 per cent, depending on the development of the emissions of other substances which are related to the decomposition of methane in the atmosphere, such as carbon monoxide and other hydrocarbons. At a conference in Toronto in 1988 a provisional guideline for the future worldwide development of CO₂ emissions was drawn up specifying reductions of 20 per cent by 2005 and 50 per cent by 2025 relative to 1988 (WMO, 1988). The Toronto guideline has meanwhile been accepted as a policy basis by a number of countries.

Since 1988, there have been further advances in the international community's response to the threat of climate change. After 15 months work, the United Nations Framework Convention on Climate Change (FCCC) was adopted by the Intergovernmental Negotiating Committee (INC) on 9 May 1992, and was opened for signature during the United Nations Conference on Environment and Development (UNCED) in Rio de Janeiro from 4 June 1992. Over 160 countries have now signed the Convention, including all major industrialised countries and leading developing countries such as India, China, Brazil and Mexico. The Convention entered into force in March 1994.

to achieve stabilisation of greenhouse gas concentrations in the atmosphere at a level that would prevent dangerous anthropogenic interference with the climate system ... within a timescale sufficient to allow ecosystems to adapt naturally to climate change, to ensure that food production is not threatened and to enable economic development to proceed in a sustainable manner.

Its main features are:

• a commitment by all parties to formulate, implement, publish and regularly update national programmes containing measures to limit emissions of greenhouse gases and protect sinks; and to report on these programmes and have them reviewed by the conference of the parties to the Convention;
• a commitment by developed countries (including the EU and its Member States) and other countries in Europe, including those undergoing the process of transition to a market economy, to take measures aimed at returning their emissions of CO₂ and of other greenhouse gases to 1990 levels by the year 2000;
• a commitment by OECD countries to provide new and additional funding to help developing countries draw up their national programmes, and to help meet the incremental costs of projects and measures which they identify in their programmes;
• a commitment by all parties to cooperate in further research into climate change and its effects, and to review regularly the working of the Convention, and the obligations to which countries have agreed.

A debate has recently started on when actions aimed at mitigating climatic change effects should be started. Some model results have been interpreted to suggest that waiting for another ten years in order to attain firmer scientific ground would imply at most a 5 per cent warmer global average temperature in 2100 than under an immediate action programme (Global Environmental Change Report, 1992). This argument may be short-sighted for several reasons:

• models do not allow for abrupt changes in the climate's response to global warming (such as shifts in ocean circulation, polar ice-cap changes or the release of methane from clathrates);
• taking only the global average as an indicator may overlook the possibilities for severe regional changes (eg, through interactions with the hydrological cycle) while the global average changes only slightly;
• non-greenhouse impacts (eg, formation of photo-oxidants, ozone depletion) may justify immediate action;
• it is unclear whether scientific knowledge or models will be more reliable after ten years and then still another ten years delay may be demanded; and
• if in any event reduction policy has to be started, it is most cost-effective to start as soon as possible in order to achieve the planned emission cuts.

If, for example, the world had immediately shifted in 1990 to a 20-year transition towards IPCC scenario B, CO₂ emissions would have had to be reduced by 7 per cent per year; waiting instead until 2000 to start the transition will require an annual CO₂ reduction of 20 per cent.

If CO₂-equivalent doubling in the next century is to be avoided, drastic reductions of global CO₂ emissions are primarily needed. In the IPCC business-as-usual scenario, an increase of 60 per cent is expected for these emissions by 2010. If no further measures are taken, only draconian reductions of 80 to 90 per cent within a few years after 2010 can postpone severe climate effects. If, by contrast, the emissions could be reduced by 2010, the most severe climate risks could be postponed beyond 2100. In that case, there would be time to halve CO₂ emissions by 2050 to reach a stabilisation level. This illustrates the need for rapid decisions.

The potential role of temperate forests as sinks for CO₂ has also been under scrutiny (see Box 23F). Germany has committed itself to a 25 to 30 per cent CO₂ emission reduction by 2005 (ie, 250 to 300 million tonnes per year), in which measures to enhance carbon sequestration by forests play an important role in the final goal. It has been estimated that expansion of forest area, a further increase of carbon storage by appropriate management, and the restoration and protection of forest health impaired by air pollution would result in an additional storage of 17 to 20 million tonnes of CO₂ per year, or 6 to 8 per cent of the reduction target. In order to store most carbon more rapidly, short rotation, fast-growing trees which provide products that last more than one rotation should be planted in soils with initially small organic carbon contents (Cannell et al, 1992). However, the role of forests as a sink for CO₂ may be minor and should be put into context since most of the planet's carbon is stored in sediments on the ocean floor.

Reductions of CO₂ emissions are technically feasible, even if consumer demands rise considerably, by increasing efficiency and by energy savings, by increased use of renewable energy sources (such as solar and wind energy) and ­ temporarily ­ by shifting fossil fuel energy generation from coal and oil to natural gas. More efficient use of nitrogen fertilisers should be stimulated to decrease N₂O fluxes. Environmental engineering solutions such as injection of carbon dioxide into the deep ocean has been shown to be theoretically feasible, but will obviously have limited application.

Because methane is a potent greenhouse gas, reduction in CH₄ emissions would be 20 to 60 times more effective in reducing the potential warming of the Earth's atmosphere over the next century than reductions in CO₂ emissions (IPCC, 1990). Reducing methane emissions will provide the additional benefits of reducing the potential for increasing tropospheric ozone (see Chapter 32). Furthermore, methane released by human activities can generally be viewed as a wasted resource, therefore reduction measures could be low cost, if not profitable (Hogan et al, 1991). It appears to be technically feasible to reduce anthropogenic CH₄ emissions by about 30 per cent (120 Tg per year). Over the next ten years, about a third to a half of these reductions would be needed to achieve the necessary 40 to 60 Tg reduction to stabilise atmospheric CH₄ concentrations. Some of the options are (Hogan et al, 1991): recovery systems in landfills; degasification before and during mining operations; improved handling and distribution in oil and natural gas systems (especially in Eastern Europe); use of specific feeds for ruminants; recovery systems for animal wastes and wastewater; improved management of rice cultivation; and fire-management and alternative agricultural practices.

Since the publication of the 1990 IPCC assessment, new evidence has emerged on the indirect effect of CFCs on global warming. By destroying ozone, the release of CFCs reduces the greenhouse effect. However, the problem is complicated by the fact that this process is strongly dependent on geographical location, especially latitude. Radiative calculations along a vertical column show that the negative effect of CFCs can cancel out the positive effect in some cases. The important conclusion is that, firstly, CFCs must be eliminated because of their role in the depletion of the stratospheric ozone layer, and, secondly, that their phase-out will not contribute significantly, if at all, to the alleviation of climate change, as previously thought.

According to the London agreement on substances that deplete the ozone layer, a phase-out of fully halogenated CFCs and halons will be achieved by the end of the century. However, envisaged substitutes will still provide some chlorine to the stratosphere and, like CFCs, have a high global warming potential. Consequently, depending on the type of halocarbons H(C)FC used to replace CFCs, implementation of the London agreement could result in even higher temperature forcing by the end of the next century. The unrestricted use of HCFCs has thus become an issue and, in a non-binding resolution, some parties of the Montreal Protocol call for a phase-out of HCFC by 2030. If such a phase-out leads to more abundant use of Class I HFCs (HFC-134a, -143a and -125), equilibrium temperature forcing by these compounds could even cause a greater warming (up to 55 per cent more) than if HCFC use were unrestricted (Kroeze and Reijnders, 1992). According to the same simulations, only a scenario involving virtual phase-out of CFCs by 2000, a phase-out of HCFC-22 by 1993, no use of Class I H(C)FCs to replace CFCs, no use of H(C)FCs for open foam, plus good housekeeping and maximum recycling or destruction of halocarbons to minimise emissions into the atmosphere, would not exceed unsustainable warming forcing of 0.08°C. It can therefore be argued that there is a case for the preference of non-climate forcing substitutes wherever possible, in order to minimise further global warming. For application such as aerosols, cleaning/drying, open and closed foam and fire extinguishers, non-H(C)FC substitutes are technically available. Most of the H(C)FCs to be used for refrigeration can probably also be replaced by non-H(C)FCs. Only for mobile air conditioning is a non-H(C)FC substitute currently unavailable.

Thus, there is a case for guiding the choice of H(C)FCs to replace CFCs by the use of both ozone depleting potentials and global warming potentials. The quest for alternatives by producers and users of halocarbons seems to be focusing on halocarbon effects on the ozone layer, so that the priority is given to HFCs over HCFCs or blends containing HCFCs. According to calculations (Kroeze and Reijnders, 1992), this results in an increase of the average global warming potential of the halocarbons used. If H(C)FCs are used without restriction to replace CFCs and halons, halocarbon applications that contribute most to temperature forcing are calculated to be refrigeration and mobile air conditioning. The gases that are expected to contribute most to calculated temperature forcing are HCFC-22 and HFC-134a. If HCFCs are phased-out to protect the ozone layer and replaced by HFCs, then HFC-134a, HFC-143a and HFC-125 will be the most important contributors to global warming, being emitted mainly from refrigeration and mobile air conditioning appliances.

International initiatives will certainly foster better harmonisation and emulations between countries in order to reach sustainable development levels. Steps towards this were taken at the UNCED in 1992 when nearly all European countries signed the Framework Convention on Climate Change. Furthermore, the European Commission is trying to reach an agreement on the use of tax instruments to stabilise CO₂ emissions at 1990 level by the year 2000.