1. The climate change - Agriculture conundrum

WIM G. SOMBROEK
Land and Water Development Division, and Interdepartmental Working Group on Climate Change, FAO, Rome, Italy

RENÉ GOMMES
Agrometeorology Group (Environmental Information Management Service), and Interdepartmental Working Group on Climate Change, FAO, Rome, Italy

The world agricultural context
The changing agricultural environment
Plant physiological direct effects
The hydrological cycle and soils
Conclusions
References

The risks associated with climate change lie in the interaction of several systems with many variables that must be collectively considered. Agriculture (including crop agriculture, animal husbandry, forestry and fisheries) can be defined as one of the systems, and climate the other. If these systems are treated independently, this would lead to an approach which is too fragmentary. The issue is more global. It is now held as likely that human activities can affect climate, one of the components of the environment. Climate in turn affects agriculture, the source of all food consumed by human beings and domestic animals. It must be further considered that not only climate may be changing, but that human societies and agriculture develop trends and constraints of their own which climate change impact studies must take into consideration.

An expert meeting held at FAO Headquarters in Rome from 7 to 10 December 1993 considered the direct effects of changing hydrological, pedological and plant physiological processes on agricultural production and concentrated on mechanisms. This introductory chapter of its Proceedings looks beyond the technical aspects of agriculture and stresses some of the major goals of FAO.

Firstly, sustainability of agricultural and rural development. How will the links between environmental resources and demography be affected in the coming 50 years? Will it be possible, at the same time, to increase food production without irremediably losing environmental resources like soils or biodiversity?

Secondly, improved food security and nutrition, two members of a spiral which also includes rural poverty and demographic pressure. What are the prospects of breaking the vicious poverty circle in many developing countries under changing climate conditions?

The world agricultural context

In general, global food production has been growing faster than human population (Table 1.1). There are, however, marked disparities between continents. For example, in Africa local production of cereals cannot keep pace with population increase, and production of root and tuber crops is growing faster than that of the nutritionally more valuable cereals.

In contrast, arable land growth lags behind population growth, which indicates some intensification of production. In Asia, the upper limit of available land has been reached in several countries, resulting in very high cropping intensities and a dominant role for irrigation. In Latin America, the increase in arable land is achieved only at a high ecological cost (especially deforestation) which may have direct relevance to climate change.

According to a recent FAO prospective study covering the years 1988/1990 to 2010 (World Agriculture: Towards 2010 or AT-2010; FAO, 1995a), 'the rate of growth of agricultural land will be further reduced during the next two decades. The pressures on fresh water resources, however, will be considerable, as will be those on the environment arising from the intensification of land use'.

A significant reduction in world population growth rates is foreseen from 1.8% (1980/1990) to 1.4% (2000/2010), roughly equivalent to an increase of the population doubling time from 40 to 50 years. The projected continuation of the high population growth in Africa can be related to slow economic development during the coming two decades, since in general a reduction of poverty precedes reduction of population growth.

AT-2010 also lists the following significant trends for the near future: (i) world production of cereals will continue to grow, but not in per capita terms; (ii) export cereals will undergo a modest growth in demand; (iii) the livestock sector in developing countries will continue to grow; (iv) root crops, tubers and plantains will retain their importance; (v) oil crops will undergo rapid growth in developing countries; and, most importantly, (vi) many developing countries will become net agricultural importers.

As indicated above for population increase, large differences exist in the rate of agricultural development among the developing continents, while differences among the developed countries tend to level out.

It is also likely that hunger and under-nutrition, which currently affects 800 million people (about 20% of the 4 000 million inhabitants of developing countries), will be reduced, but large pockets of malnutrition will persist. Pressure on environmental resources will continue to build up.

Table 1.1. Some trends in worldwide population, food production, arable and irrigated land (data from FAO, 1990). Europe and Asia do not include the former USSR. Arable land is dermed as all agricultural land, excluding perennial crops and permanent rangeland. Irrigated land is in % of the land under annual and permanent crops, average value from 1981 to 1990. Fertilizer use is the 1980-1990 average (after Gommes, 1993)

Continent	1961-1990 exponential growth rate (%)					% Cereal yield increase (1961-1970 and 1986-1990)	Fertilizer use (kg fertilizer per ha arable land)	Irrigated land
	Population	Production			Arable land
		Cereals	Tubers	Pulses
Africa	2.83	1.95	2.74	1.58	0.44	32	1.2	5.8
North and Central America	1.52	2.32	1.30	1.53	-0.59	49	10.7	9.6
South America	2.34	2.80	0.51	0.71	1.53	44	2.6	5.8
Asia	2.11	3.30	1.95	0.26	0.07	86	14.5	31,0
Europe	0.51	2.37	-1.27	1.57	-0.53	75	64.3	10.8
Oceania	1.72	3.14	1.62	13.76	1.23	34	2.0	4.1
World	1.88	2.66	0.78	0.83	0.12	65	9.6	15.2

The changing agricultural environment

THE CLIMATE 'COMPLEX'

Climate constitutes a complex of inter-related variables. On average, through a set of regulatory mechanisms, a smooth change in one variable triggers smooth changes in most others. With the exception of possible qualitative and abrupt variations, which will be mentioned below, such inter-relations are independent of atmospheric carbon dioxide (CO₂). The latter and other greenhouse gases play a part largely through their effect on the radiation balance of the atmosphere.

There is only a weak link between such factors as cloudiness and wind. Temperature, evaporation and rain are strongly correlated, which illustrates the likely intensification of the hydrological cycle (Figure 1.1). Combined with the projected pressure on land and water use, competition for land and water will certainly become a key social and political issue.

Climate variability is likely to increase under global warming (Katz and Brown, 1992), both in absolute and in relative terms. This is linked with thresholds which affect the occurrence of many meteorological phenomena. For instance, tropical cyclones are 'fed' by water vapour evaporating from oceans at a temperature above 26 or 27°C. Therefore, higher average sea surface temperatures are bound to result in a higher frequency of tropical cyclones.

The rate of change itself is extremely important. For example, recent work (Lehman, 1993; Paillard and Labeyrie, 1994; Rahmsdorf, 1994; Holmes, 1995) on the saw-tooth temperature changes in the past as observed in the Arctic, raise concern that changes may occur abruptly, with average temperatures changing by 10 or 12°C in just a matter of decades. The mechanism of such changes is not clear as yet, but seems to involve the mechanical stability of ice sheets, and sudden changes in the hydrological cycle and the Atlantic conveyor current.

Figure 1.1. Some relationships between major climate variables (average temperature, Temp.; water vapour pressure, Vap.; rainfall, Rain; wind speed, Wind; cloudiness, Clouds; and evapo-transpiration potential, ETP). Solid and dotted lines indicate positive and negative correlations, respectively. The strength of the correlation decreases from double heavy lines to thin single lines. Computed from annual averages of 3 263 land stations, mostly from developing countries (FAO, 1995b)

Such changes would, of course, be associated with dramatic changes in the distribution and quantities of ocean products, and cause havoc to established national fishery activities. They would also make adaptation to climate change, together with most agricultural planning, extremely difficult.

AGRICULTURAL GREENHOUSE GASES

In addition to water vapour, important greenhouse gases are carbon dioxide (CO₂), methane (CH₄), nitrous oxide (N₂O), tropospheric ozone (O₃) and chlorofluorocarbons (CFCs). The basic characteristics of the first three gases are given in Table 1.2. The degree to which these greenhouse gases stem from agricultural sources is also given. The exact ratio between these land-use related emissions and those from natural ecosystems (swamps, tundra), fossil fuels (coal, oil, gas) and geological sources (volcanoes) are a bone of contention among industrialized, oil producing and developing countries.

Moral rights and duties in relation to the international flow of development and conservation funds are involved. Therefore, better estimates based on exact measurements of the net greenhouse gas emissions from agricultural practices in developing countries are urgently required to assess responsibilities properly. The required reductions of emissions to achieve stabilization of atmospheric concentrations of current levels are believed to be >60% for CO₂ 15-20% for CH₄ and 70-80% for N₂O. This implies the need for up-to-date and complete information on land cover and land uses per country: not only the kind of crops grown but also their intensity, their rotation and the amount of inputs (energy, fertilizers).

Table 1.2. An overview of agricultural greenhouse gases with the trends as currently envisaged (adapted from IPCC, 1992; Houghton et al., 1995; Keeling et al., 1995). ppmv and ppbv stand for parts per 10⁶ and parts per 10⁹, respectively, by volume.

	CO2	CH4	N2O
Atmospheric lifetime (yr)	120	14.5	120
Direct GWP 1	1	24.5 2	320
Pre-industrial concentration 3	280 ppmv	0.8 ppmv	288 ppbv
Present-day levels	360 ppmv	1.72 ppmv	310 ppbv
Current annual increase (%)	0.5	0.9	0.25
Major agricultural sources 4	deforestation	- wetland rice	- synthetic N fertilizers
		- ruminants	- animal excreta
		- biomass burning	- biological N fixation
Percentage of global source stemming from agriculture	30	40	25
Predicted change 1990-2020	-	+	+

¹ GWP, Global Warming Potential, is the direct warming effect in relation to CO₂ at a time horizon of 100 years.

² Includes indirect effects through chemistry.

³ About year 1750-1800.

⁴ Activities responsible for emissions are projected to increase by: rice (+ 10%); ruminant population (+30%); synthetic fertilizer use (+20%): animal excreta and biological N fixation increasing but rate not specified.

The current trends of some of the agricultural sectors directly associated with greenhouse gas emission sources are listed in Table 1.3.

The loss of 'natural land', including tropical forests to agriculture, grazing, logging and urbanization, may continue, though at a slower pace (AT-2010). Many earlier estimates are now regarded with suspicion, as the borderline between crop agriculture, forest and cattle agriculture appears to be a fuzzy one.

Finally, both recent trends of fertilizer use and AT-2010 projections indicate major increases in consumption. Even with appropriate measures to optimize fertilizer use, it is likely that N₂O losses from fertilizers will continue to increase.

Ecological and indirect climate effects

In qualitative terms, many indirect effects of climate change on agriculture can be conjectured. Most of them are estimated to be negative and they catch most of the attention of the media. These effects include:

· the overall predictability of weather and climate would decrease, making the day-to-day and medium-term planning of farm operations more difficult;

· loss of biodiversity from some of the most fragile environments, such as tropical forests and mangroves;

· sea-level rise (40 cm in the coming 100 years) would submerge some valuable coastal agricultural land;

· the incidence of diseases and pests, especially alien ones, could increase;

· present (agro) ecological zones could shift in some cases over hundreds of kilometres horizontally, and hundreds of metres altitudinally, with the hazard that some plants, especially trees, and animal species cannot follow in time, and that farming systems cannot adjust themselves in time;

· higher temperatures would allow seasonally longer plant growth and crop growing in cool and mountainous areas, allowing in some cases increased cropping and production. In contrast, in already warm areas climate change can cause reduced productivity;

· the current imbalance of food production between cool and temperate regions and tropical and subtropical regions could worsen.

Table 1.3. Growth rates between 1961 and 1990 in agricultural sectors responsible for greenhouse gas emissions (from FAO, 1990). Europe and Asia do not include the former USSR. Domestic ruminant numbers were computed as the sum of cattle, sheep, goats, camels and buffaloes

Continent	1961-1990 exponential growth rate (%)
	Ruminant numbers	Forested area	Rice area	Fertilizer consumption
Africa	1.29	-0.43	2.23	6.21
N and C America	-0.07	-0.02	1.50	3.29
S. America	1.29	-0.49	1.65	9.15
Asia	1.18	-0.59	0.62	9.54
Europe	0.33	0.25	0.78	2.75
Oceania	0.07	-1.15	5.88	1.25
World	0.90	-0.26	0.74	5.35

Plant physiological direct effects

The greenhouse gases CH₄, N₂O and chlorofluorocarbons (CFCs) have no known direct effects on plant physiological processes. They only change global temperature and are therefore not discussed further. Instead, concentration should be on the effects of increased CO₂ tropospheric O₃, increased UV-B through depleted stratospheric ozone, increased temperatures and the associated intensification of the hydrological cycle.

CARBON DIOXIDE

The CO₂ fertilization effect

CO₂ is an essential plant 'nutrient', in addition to light, suitable temperature, water and chemical elements such as N, P and K, and it is currently in short supply.

Higher concentrations of atmospheric CO₂ due to increased use of fossil fuels, deforestation and biomass burning, can have a positive influence on photosynthesis (Figure 1.2); under optimal growing conditions of light, temperature, nutrient and moisture supply, biomass production can increase, especially of plants with C3 photo-synthetic metabolism (Box 1.1), above and even more below ground (for details see chapter 4 by Allen et al.).

Figure 1.2. Schematic effect of CO₂ concentrations on C₃ and C₄ plants (after Wolfe and Erickson, 1993). The main mechanism of CO₂ fertilization is that it depresses photo-respiration, more so in C₃ than in C₄ plants

Box 1.1. The major agricultural crops and the three photosynthetic pathways Plants are classified as C3, C4 or CAM according to the products formed in the initial phases of photosynthesis. C3 species respond more to increased CO2; C4 species respond better than C3 plants to higher temperature and their water-use efficiency increases more than for C3 plants. There are some indications that enhancements can decline over time ('down-regulation') (see chapter by Allen [chapter 4 in this book]). C3 plants: cotton, rice, wheat, barley, soybeans, sunflower, potatoes, most leguminous and woody plants, most horticultural crops and many weeds C4 plants: maize, sorghum, sugar cane, millets, halophytes (i.e., salt-tolerant plants) and many tall tropical grasses, pasture, forage and weed species CAM plants (Crassulacean Acid Metabolism, an optional C3 or C4 pathway of photosynthesis, depending on conditions): cassava, pineapple, opuntia, onions, castor

A total of 10 to 20% of the approximate doubling of crop productivity over the past 100 years could be due to this effect (Tans et al., 1990) and forest growth or regrowth may have been stimulated as well. Further productivity increases may occur in the coming century, in the order of 30% or more where plant nutrients and moisture are adequate.

Higher CO₂ values would also mitigate the plant growth damage caused by pollutants such as NO_x and SO₂ because of smaller stomatal openings (see below). Higher percentages of starch in grasses improves their feeding quality, implying less need for feed mixes when silaging.

The CO₂ anti-transpirant effect (improved water-use efficiency WUE)

With increased atmospheric CO₂ the consumptive use of water becomes more efficient because of reduced transpiration. This is induced by a contraction of plant stomata and/or a decrease in the number of stomata per unit leaf area. This restricts the escape of water vapour from the leaf more than it restricts photosynthesis (Wolfe and Erickson, 1993; see details in chapter by Van de Geijn and Goudriaan).

With the same amount of available water, there could be more leaf area and biomass production by crops and natural vegetation. Plants could survive in areas hitherto too dry for their growth.

ULTRAVIOLET RADIATION

Increased ultraviolet radiation (UV-B, between 280 and 320 nanometres), due to depletion of the stratospheric ozone layer, mainly in the Antarctic region, may negatively affect terrestrial and aquatic photosynthesis and animal health. Over the last decade, a decrease of stratospheric ozone was observed at all latitudes (about 10% in winter, 0% during summer and intermediate values during spring and autumn). However, the 'Biological Action Factor' of UV-B can vary over several orders of magnitude with even slight changes in the amount and wavelength of UV-B.

The subject is treated in detail elsewhere in this book (see chapter by Unsworth and Hogsett, or Runeckles and Krupa, 1994). In particular it can be noted that:

· there are damaging effects of increasing UV-B on crops, animals and plankton growth. It has been reported that UV-B affects the ability of plankton organisms to control their vertical movements and to adjust to light levels;

· reductions in yield of up to 10% have been measured at experimentally very high UV-B values, and would be particularly effective in plants where the CO₂ fertilization effect is strongest. On the other hand, UV-B increase could increase the amount of plant internal compounds that act against pests.

TROPOSPHERIC OZONE

Tropospheric ozone originates about half from photochemical reactions involving nitrogen oxides (NO_x), methane or carbon monoxide, and half by downward movement of stratospheric ozone.

High ozone concentrations have toxic effects on both plant and animal life (German Bundestag, 1991; MacKenzie and El-Ashry, 1988; UNEP, 1993). It is likely that ozone, in conjunction with other photo-oxidants, is contributing towards the 'new type of forest damage' observed in Europe and the United States.

In the tropics, tropospheric ozone concentrations are generally lower than at northern mid-latitudes. However, this does not apply to periods when biomass burning releases precursor substances for the photochemical formation of ozone.

TEMPERATURE

In general, higher temperatures are associated with higher radiation and higher water use. It is relatively difficult to separate the physiological effects (at the level of plants and plant organs) of temperatures from the ecological ones (at the level of the field or of the region). There are both positive and negative impacts at the two levels, and only crop- and site-specific simulation can assess the global 'net' effect of temperature increases (for details see chapter by Abrol and Ingram). It is generally agreed that:

· rising temperatures - now estimated to be 0.2°C per decade, or 1 °C by 2040 (Mitchell et al., 1995) with smallest increases in the tropics (IPCC, 1992) - would diminish the yields of some crops, especially if night temperatures are increased (the temperature increase since the mid-1940s is mainly due to increasing night-time temperatures, while CO₂-induced warming would result in an almost equally large rise in minimum and maximum temperatures (Kukla and Karl, 1993);

· higher temperatures could have a positive effect on growth of plants of the CAM type. They would also strengthen the CO₂ fertilization effect and the CO₂ anti-transpirant effect of C₃ and C₄ plants (see Box 1.1) unless plants get overheated;

· higher night temperature may increase dark respiration of plants, diminishing net biomass production;

· higher cold-season temperatures may lead to earlier ripening of annual crops, diminishing yield per crop, but would allow locally for the growth of more crops per year due to lengthening of the growing season. Winter kill of pests is likely to be reduced at high latitudes, resulting in greater crop losses and higher need for pest control;

· higher temperatures will allow for more plant growth at high latitudes and altitudes.

COMBINED EFFECTS AND SOME UNCERTAINTIES

The changes in CO₂ tropospheric ozone and increased UV-B do not necessarily occur simultaneously: CO₂ increase is worldwide, but with a strong seasonality in middle and higher latitudes; significant increase of UV-B is largely limited to subpolar regions (and mainly during the northern hemisphere winter months); high near-surface O₃ levels are restricted to the neighbourhood of major cities, airports, etc. (Seitz, 1994)

Box 1.2 illustrates some of the potential mechanisms which could account for either increased or decreased biomass under global change conditions. Note that increased biomass could even be associated with a decreased yield of grain (or sugar, oil, etc.) if one of the consequences of increased CO₂ will be a redistribution of biomass among plant organs (there are indications of relative increases of root growth).

The hydrological cycle and soils

Even a slight increase in surface temperatures will affect evaporation, atmospheric moisture and precipitation (Figure 1.1). While it is generally agreed that rainfall will increase (by an estimated 10 to 15%), two aspects have to be elucidated: how will rainfall intensities be affected, and what are the details of spatial changes. This is still largely a matter of discussion among experts (for details see chapter by Evans).

Based on palaeoclimatic analogies, certain authors predict more favourable rainfall conditions in the present-day Sahel (Petit-Maire, 1992). If the increase in precipitation should be associated with increased rainfall intensities, then the quality and quantity of soil and water resources would decline, for instance through increased runoff and erosion, increased land degradation processes, and a higher frequency of floods and possibly droughts. However:

· the extra precipitation on land, if indeed including present subhumid to semi-arid areas, will increase plant growth in these areas, leading to an improved protection of the land surface and increased rainfed agricultural production; in already humid areas the extra rainfall may, however, impair adequate crop drying and storage;

· the extra precipitation predicted to occur in some regions provides possibilities for off-site extra storage in rivers, lakes and artificial reservoirs (on-farm or at subcatchment level) for the benefit of improved rural water supply and expanded or more intensive irrigated agriculture and inland fisheries:

· the effects on water resources and water apportioning of international river and lake basins can be very substantial, with political overtones.

Box 1.2. Some mechanisms likely to affect biomass production under global change conditions. Note that the ratio between economic yield (e.g., grain, fibre) and biomass may change relative to current conditions
ETP: Evapotranspiration potential	WHC: Soil water holding capacity
ETA: Actual evapotranspiration	OM: Organic matter
WUE: Water-use efficiency	LAI: Leaf area index
The heavy line indicates a hypothetical link between increased humidity and cloudiness.

The greatest risks are often estimated to be associated with increased soil loss through erosion.

Soils, as a medium for plant growth, would be affected in several other ways (for details see chapter by Brinkman and Sombroek):

· increased temperatures may lead to more decomposition of soil organic matter;

· increased plant growth due to the CO₂ fertilization effect may cause other plant nutrients such as N and P to become in short supply; however, CO₂ increase would stimulate mycorrhizal activity (making soil phosphorus more easily available), and also biological nitrogen fixation (whether or not symbiotic). Through increased root growth there would be extra weathering of the substratum, hence a fresh supply of potassium and micronutrients;

· the CO₂ fertilization effect would produce more litter of higher C/N ratio, hence more organic matter for incorporation into the soil as humus; litter with high C/N decomposes slowly and this can act as a negative feedback on nutrient availability;

· the 'CO₂ anti-transpirant' effect would stimulate plant growth in dryland areas, and more soil protection against erosion and lower topsoil temperatures, leading to an 'anti-desertification effect'.

Conclusions

Global climate change, if it occurs, will definitely affect agriculture. Most mechanisms, and two-way interactions between agriculture and climate, are known, even if not always well understood.

It is evident that the relationship between climate change and agriculture is still very much a matter of conjecture with many uncertainties (see also Rosenzweig and Hillel, 1993); it remains largely a conundrum.

Major uncertainties affect both the Global Circulation Models (GCMs) and the response of agriculture, as illustrated by differences among models, especially as regards effects at the national and subregional levels. In addition, many of the models do not take into consideration CO₂ fertilization and improved water-use efficiency, the effect of cloud cover (on both climate and photosynthesis), or the transient nature of climate change.

It is also worth remembering that enormous knowledge gaps still affect the carbon cycle (with a missing sink of about 2 Gt of carbon), the factors behind the recent near-stabilization of the atmospheric methane concentrations or the unexplained reduced rate of CO₂ increase in recent years, the effect of volcanic eruptions (such as the recent Pinatubo eruption), the effect of any increased cloudiness, etc.

References

FAO. 1990. 1989 Production Yearbook No. 43. FAO Statistics Series No. 94. AGROSTAT-PC, the PC version of the yearbook, was also used.

FAO. 1995a. World Agriculture: Towards 2010, An FAO Study. N. Alexandratos (ed.). John Wiley, Chichester, UK, and FAO, Rome. 488 p.

FAO, 1995b. FAOCLIM 1.2, Worldwide agroclimatic data. FAO Agrometeorology Working Paper Series No. 11. FAO, Rome. 1 CD-ROM and 66 p.

German Bundestag. (ed.). 1991. Protecting the Earth. A status report with recommendations for a new energy policy. Bonn, Deutscher Bundestag, Referat Oeffentlichkeitsarbeit. Vol. 1, 672 p.

Gommes, R. 1993. Current climate and population constraints on world agriculture. In: Agricultural Dimensions of Global Climate Change. H.M. Kaiser and T.E. Drennen (eds.). pp. 67-86.

Holmes, R. 1995. Arctic ice shows speed of climate 'flips'. New Scientist 145 (1967): 13.

Houghton, J.T., Meira Filho, L.G., Bruce, J., Lee, H., Callander, B.A., Haites, E., Harris, N. and Maskell, K. (eds.). 1995. Climate Change 1994. Radiative forcing of climate change; and an evaluation of the IPCC IS92 emission scenarios. Cambridge University Press, Cambridge, New York, Melbourne.

IPCC. 1992. Climate Change 1992. The supplementary report to the IPCC scientific assessment. J.T. Houghton, B.A. Callandar and S.K. Varney (eds.). Cambridge University Press. Cambridge.

Kaiser, H.M. and Drennen T.E. (eds.). 1993. Agricultural Dimensions of Global Climate Change. St. Lucie Press, Delray Beach, Florida. 311 p.

Katz, R.W. and Brown, B.G. 1992. Extreme events in a changing climate: variability is more important than averages. Clim. Change 21: 289-302.

Keeling, C.D., Whorf, T.P, Wahlen, M. and van der Plicht, J. 1995. Interannual extremes in the rate of rise of atmospheric carbon dioxide since 1980. Nature 375: 666-670.

Kukla, G. and Karl, T.R. 1993. Nighttime warming and the greenhouse effect. Envir. Sci. Technol. 27 (8): 1468-1474.

Lehman, S. 1993. Ice sheets, wayward winds and sea change. Nature 365: 108-110.

MacKenzie, J.J. and El-Ashry, M.T. 1988. Links between air pollution and damage to crops. In: Ill Winds: Airborne Pollution's Toll on Trees and Crops. World Res. Institute. pp. 25-33.

Mitchell, J.F.B., Johns, T.C., Gregory, J.M. and Tett, F.B. 1995. Climate response to increasing levels of greenhouse gases and sulphate aerosols. Nature 376: 501-504.

Paillard, D. and Labeyrie, L. 1994. Role of thermohaline circulation in the abrupt warming after Heinrich events. Nature 372: 162-164.

Petit-Maire, N. 1992. Lire l'avenir dans les archives géologiques. La recherche 23 (243): 566-569.

Rahmsdorf, S. 1994. Rapid climate transitions in a coupled ocean-atmosphere model. Nature 372: 82-85.

Rosenzweig, C. and Hillel, D. 1993. Agriculture in a greenhouse world: potential consequences of climate change. National Geographic Research and Exploration 9: 208-221.

Runeckles, V.C. and Krupa, S.V. 1994. The impact of UV-B radiation and ozone on terrestrial vegetation. Envir. Pollut. 83 (1-2): 191-213.

Seitz, F. 1994. Global Warming and Ozone Hole Controversies, A Challenge to Scientific Judgement. G.C. Marshall Institute, Washington DC. 32 p.

Solbrig, O.T., van Emden, H.M. and Van Oordt, P.G.W.J. 1992. Biodiversity and global change. IUBS Monograph No. 8, IUBS Press, Paris. 222 p.

Tans, P.P., Fung, I.Y. and Takahashi, T. 1990. Observational constraints on the global atmospheric CO₂ budget. Science 247: 1431-1438.

UNEP. 1993. Action on Ozone. UNEP, Nairobi. 24 p.

Wolfe, D.W. and Erickson, J.D. 1993. Carbon dioxide effects on plants: uncertainties and implications for modelling crop response to climate change. In: Agricultural Dimensions of Global Climate Change. H.M. Kaiser and T.E. Drennen (eds.). pp. 153-178.