Potential Impact of Climate Change on Agriculture and Food Supply

David W. Wolfe
Cornell University

Will climate change help or hinder our efforts to maintain an adequate food supply for the increasing world population of the next century? Which regions are likely to benefit and which are likely to suffer food shortages and socioeconomic crises? Could the beneficial effects of increasing atmospheric carbon dioxide (CO2) on plants (the so-called "CO2 fertilization effect") counteract some of the negative effects of climate change? What types of adaptations and policies will be necessary to take advantage of the opportunities and minimize the negative impacts of climate change on agriculture? What will the cost of these adaptations and policies be?

To address these questions scientists from various disciplines have linked together climate, crop growth, and economic-food trade computer models. These multi-layered models are extremely complex and contain numerous assumptions about the physical, biological, and socioeconomic systems they attempt to simulate. Nevertheless, they represent the most comprehensive analyses we have at present. They can be useful to policymakers, particularly if there is an educated appreciation for the level of uncertainty inherent in their projections. Before presenting model outcomes, we will first review some fundamental aspects of what we know and don't know about how crop plants respond to temperature and increases in atmospheric CO2.

Temperature Effects on Plants

Most plant processes related to growth and yield are highly temperature dependent. We can identify an optimum temperature range for maximum yield for any one crop. Crop species are often classified as warm- or cool-season types. The optimum growth temperature frequently corresponds to the optimum temperature for photosynthesis, the process by which plants absorb CO2 from the atmosphere and convert it to sugars used for energy and growth. Temperature also affects the rate of plant development. Higher temperatures speed annual crops through their developmental phases. This shortens the life cycle of determinate species like grain crops, which only set seed once and then stop producing. Figure 1 illustrates the temperature effects on photosynthesis and crop growth duration. It shows that for a variety currently being grown in a climate near its optimum, a temperature increase of several degrees could reduce photosynthesis and shorten the growing period. Both of these effects will tend to reduce yields.

The particular crop varieties currently being grown in major production areas are usually those best-adapted to the current climate. A significant increase in growing season temperatures will require shifts to new varieties that are more heat tolerant, do not mature too quickly, and have a higher temperature optimum for photosynthesis. Developing such varieties should be possible for many crop species, but there are limits to what can be accomplished through plant breeding and modern genetic engineering approaches. In many cases traditional crops will have to be abandoned for new crops better suited to the new environment. For farmers in very cool regions, where the current climate limits their crop options, global warming will be mostly a benefit, giving them the opportunity to grow a wider range of crops and long-growing-season, high-yielding varieties.

Some plant species require a cold period before they will produce flowers and a harvestable product. The process, called vernalization, tends to have very narrow temperature and duration boundaries. Vernalization of winter wheat, for example, requires temperature to be between 0 and 11°C (32 and 52°F), with an optimum near 3°C (37°F) for a period of 6 to 8 weeks. Production of the seed of many biennial vegetable crops has similar requirements. Even a minor climate shift of 1-2°C could have a substantial impact on the geographic range of these crops.

Temperature Effects on Livestock

Climate change will affect livestock production indirectly by its impact on the availability and price of animal feed. Farm animals are also directly affected by temperature. Figure 2 illustrates that animal species differ in their temperature optimum range. Young animals have a very narrow and specific temperature optimum. A rise in temperatures in regions currently near the threshold of the optimum range could be detrimental to production. Construction and maintenance of controlled environment facilities to house farm animals is costly and will not be a viable option for many.

Carbon Dioxide (CO2) Effects on Plants

The debate over whether CO2 and other greenhouse gases are warming the planet continues, but few question the fact that atmospheric CO2 is increasing exponentially and will likely double (to 700 parts per million (ppm)) within the next century. This has a potential beneficial effect on the Earth's plant life because plants take up CO2 via photosynthesis and use it to produce sugars and grow. If this "CO2 fertilization effect" is large, it could significantly increase the capacity of plant ecosystems to absorb and temporarily store excess carbon. It could also lead to significant increases in crop productivity.

CO2 and photosynthesis

The biochemistry of photosynthesis differs among plant species, and this greatly affects their relative response to CO2. Most economically important crop and weed species can be classified as either a C3 or C4 type, the names referring to whether the early products of photosynthesis are compounds with three or four carbon atoms. It has been well known for many years that the C3 photosynthetic pathway is less efficient than the C4 pathway. Because of this, C3 plants benefit much more from increases in CO2 than C4 plants (Fig. 3). Over 90 percent of the world's plant species are the C3 type, including wheat, rice, potato, bean, most vegetable and fruit crops, and many weed species. However, the C4 group includes the important food crops, maize, millet, sugarcane, and sorghum, as well as many pasture grasses and weed species. These C4 crops will benefit little from a CO2 doubling.

The CO2 response curves shown in Figure 3 are typical of results from experiments where plants are grown under optimal conditions and current ambient CO2 concentration (350 ppm). The magnitude of the CO2 response often changes when plants are acclimated to a high CO2 environment (e.g., 700 ppm). Usually the beneficial effects decline with long-term exposure to high CO2, but in some instances they increase. Identifying the mechanisms of both upward and downward photosynthetic acclimation to CO2 is an important area of current research. The greatest benefits from CO2 tend to occur when plants are able to expand their "sink capacity" for the products of photosynthesis by, for example, producing more flowers and fruit when grown at high CO2. When genetic or environmental factors limit growth and sink capacity, sugars build up in the leaves, a negative feedback on photosynthesis occurs and the benefits from elevated CO2 become minimal. This explains why maximum CO2 benefits usually require an optimum environment and increased inputs of water and fertilizers. A research priority in the future will be breeding for genotypes that take full advantage of increases in CO2.

CO2 and crop water use

Another important direct effect of high CO2 on plants (both C3 and C4 species) is a partial closure of the small pores, or stomates, of the leaves. This restricts the escape of water vapor from the leaves (transpiration) more than it restricts photosynthesis. Some have suggested that this will moderate the increase in crop water requirements anticipated to occur in a global warming scenario. However, significant water savings have seldom been observed in experiments designed to test the hypothesis. Although plants grown at high CO2 transpire less water per unit leaf area, their leaves are frequently larger and there are more of them so that whole plant water use is similar to or greater than plants grown at normal CO2 concentrations.

The stomatal response to CO2 does appear to have some beneficial impact under water-limited conditions. Several studies have found that the relative beneficial effect of a CO2 doubling on growth is greater under mild water stress conditions than when water supply is optimum. The absolute benefit from CO2 is nevertheless maximum when water is not limiting growth.

CO2 and crop yield

Most of our information regarding the yield response to CO2 is based on controlled environment experiments, where plants were well supplied with water and nutrients, temperatures were near optimum, and pressure from weeds, disease and insect pests were nonexistent. Under such optimum conditions a doubling of CO2 (e.g., from 350 to 700 ppm) typically increases the yield of C3 crops by 20 - 35%. While this describes the average, there are reports in the literature of lower yield responses in some slow- growing winter vegetables such as cabbage, and reports of higher yield responses in some fast-growing indeterminate species such as cotton and citrus. Maize and other C4 crops typically have yield increases of less than lO% with a CO2 doubling, as might be expected from their photosynthetic response (Fig. 3).

Several recent reviews have emphasized that when plants grow in a field situation, the optimum conditions required for full realization of the benefits from CO2 enrichment are seldom, if ever, maintained. This is particularly true for natural ecosystems and for agricultural systems in developing nations where irrigation, fertilizer, herbicides, and pesticides are not available or prohibitively expensive. Even in developed countries, the increase in inputs sometimes necessary for maximum CO2 benefit may not be cost-effective or may be limited by concerns regarding resource conservation or environmental quality.

The specific temperature range for realization of a positive CO2 effect varies, but for most crops the beneficial effects become minimal at temperatures below about 15°C (59°F). This has important implications for temperate regions of the world where, despite global warming, average temperatures during early and late portions of the growing season will be too cool to expect much benefit from a CO2 doubling. Where excessively high temperatures (e.g. above 38°C or 100°F) occur, flowering and pollination of many crop species will be impaired. Yields will be very low regardless of atmospheric CO2 levels.

CO2 and weed, disease, and insect pests

An increase in atmospheric CO2 is just as likely to increase the growth rate of weed species as cash crops. To date, most of our information regarding crop response to CO2 is based on experiments in which competition from weeds was not a factor. Important C4 crops, such as maize and sugarcane, may experience yield reductions because of increased competition from C3 weeds. However, broad generalizations regarding CO2 enrichment effects on crop-weed competition provide little insight into the specific weed control challenges that farmers will have to face in the coming century. The site-specific mix of weed and crop species, and the relative response of each of these species to environmental conditions in the future CO2-rich world, will determine the economic outcome for both farmers and consumers.

Recent research examining the effect of elevated CO2 on insect damage has found that leaf-feeding insects often must consume more foliage to survive on high CO2-grown plants, presumably because the leaves tend to have a lower protein concentration. Natural selection would tend to favor the evolution of insect genotypes that consume more plant material more rapidly. To combat this, farmers may be required to use more pesticides.

The climate changes that result from increased atmospheric CO2 concentrations will undoubtedly influence the geographic range of insect and disease pests. Warmer temperatures in high latitude areas may allow more insects to overwinter in these areas. Also, crop damage from plant diseases is likely to increase in temperate regions because many fungal and bacterial diseases have a greater potential to reach severe levels when temperatures are warmer or when precipitation increases.

Model Projections of Climate Change Impact on Food Supply

Global assessment

Several groups have attempted to model the impact of climate change on crop yields and food supply. A particularly comprehensive approach, involving the collaboration of many scientists worldwide, is that described by Rosenzweig and Parry (1993, 1994). They linked global climate model outputs with crop growth models, and then used the yield projections as inputs into a world food trade model. The analysis considered climate uncertainties by comparing results from three different general circulation models (GCMs), those from the NASA Goddard Institute for Space Studies (GISS), the Geophysical Fluid Dynamics Laboratory (GDFL) and the UK Meteorological Office (UKMO). Two levels of farmer adaptation to climate change, and the potential direct effects of CO2 were also considered.

Table 1 shows estimated yield changes for several important cereal crops. It is clear that regardless of GCM used, climate change had a substantial negative effect on yield unless a beneficial effect from elevated CO2 is assumed. With a CO2 fertilization effect, the impact of climate change on wheat and soybeans shifts from negative to positive for the GISS and GDFL climate scenarios, and rice yields also significantly improve. Maize yields remain substantially negative because this C4 crop is assumed to benefit little from an equivalent CO2 doubling.

The CO2 effect was incorporated into the crop models as a simple multiplier, increasing predicted yields of soybeans, wheat, rice, and maize by 34, 22, 19, and 7%, respectively. To obtain these values, the modelers had to rely on the published literature, which, as discussed above, is dominated by experiments conducted under optimum conditions. The CO2 effect multiplier was applied without taking into account the likely profound effect of regional, seasonal, and daily temperature variations on the magnitude of CO2 response. The predicted yields with CO2 effects in Table 1, therefore, probably overestimate the yield response to CO2 in many cases. On this point the authors themselves warn "... there is always uncertainty regarding whether experimental results will be observed in the open field conditions likely to be operative when farmers are managing crops." (Rosenzweig and Parry, 1993. p. 92).

There was considerable regional variation in yield response to climate change as indicated in Table 2 which shows predicted wheat yields for several countries. Yield increases in the highest latitude locations, for example in parts of Canada and the former USSR, were due to an extension of the frost-free growing season and improved temperatures for productivity. The optimistic assumptions regarding effects of CO2 also caused substantial yield benefits in all areas. Decreased yields were associated with faster plant development rates that shortened the growing period, decreases in water availability, and poor vernalization of winter wheat.

Many developing nations are in tropical or subtropical areas where a global warming will be of little benefit, and often a detriment, to crop yields. Also, these areas frequently have less capacity for irrigation, which becomes a more serious drawback in some climate scenarios where rainfall does not meet the increasing crop water requirements. These factors, combined with a relatively high population growth rate, will tend to increase the probability of food shortages in many developing nations. Figure 4 illustrates the large discrepancy in predicted cereal production for developed vs. developing nations. This particular analysis assumed direct positive CO2 effects, and "Level 1" adaptation to climate change by farmers, which included shifts in planting date, additional irrigation in areas with existing irrigation capacity, and changing to better adapted crop varieties from the existing pool of varieties available. Figure 5 shows the predicted increase in risk of world hunger with climate change and various levels of farmer adaptation. Adaptation Level 2 included development of new irrigations systems, increase in fertilizer application, and development of new varieties. This analysis shows that only if we assume Level 2 farmer adaptation capacity and an optimistic beneficial effect of elevated CO2 on yields are the risks of hunger likely to not increase beyond what we would expect without climate change.

U.S. agriculture

Most analyses have concluded that although there will be both winners and losers within the U.S. agricultural sector, overall productivity is not likely to decline to the point of threatening national food security unless climate change is severe. The magnitude and direction of the predicted impact on U.S. agriculture varies depending on assumptions about climate and plant response to CO2. For example, the comprehensive study by Adams et al. (1990) predicted a 9% increase in production of field crops (included wheat, soybean, sorghum, cotton, oats, maize, hay and silage) using the GISS climate scenario, or a 20% yield decrease with the GFDL climate model. The GFDL model predicts slightly warmer, drier conditions for some regions of the U.S. than the GISS model. This assessment assumed an optimistic 20 - 35% yield boost due to the CO2 fertilization effect. The economic component of their analysis predicted that for the more severe GFDL climate scenario, U.S. consumers would face moderately higher food prices, but the major impact would be on exports. Both climate scenarios predicted significant reductions in cropped acreage in the Southeast, Southern Plains, and Northeast, and increases in the Northern Plains, Great Lakes, and Rocky Mountain regions.

Can Farmers Adapt to Climate Change?

Farmers in developing nations will be least able to adapt to climate change because of a relatively weak agricultural research base, poor availability of inputs such as water and seed of new varieties, and inadequate capital for making adjustments at both the farm and national level.

In contrast, the U.S. and many other developed nations have a strong agricultural research base, abundant natural resources for flexibility in cropping patterns, and capital available to pay for adaptations and buffer negative economic effects during transition. For this reason many are optimistic that farmers in developed nations will be able to take advantage of opportunities and minimize negative effects associated with climate change.

Adapting to climate change will be costly, however. Costs at the farm level will include such things as increased use of water, fertilizer and pesticides to maximize beneficial effects of higher CO2, and investment in new farm equipment and storage facilities as shifts are made to new crop varieties and new crops. Costs at the national level will include substantial diversion of agricultural research dollars to climate change issues, and major infrastructure investments, such as construction of new dams and reservoirs to meet increased crop water requirements. Environmental costs associated with agricultural expansion into some regions could include increased soil erosion, increased risk of ground and surface water pollution, depletion of water resources, and loss of wildlife habitat.

Developed as well as developing nations must be prepared to deal with the citizens in those regions negatively impacted by climate change. Regardless of capital availability, agricultural economies in some areas will collapse due to factors such as excessively high temperatures, severe pest pressure, lack of locally adapted varieties or poor markets for adapted crops. As climatic zones shift, there will be some cases where those zones with the best climate for crops will not have good soils or available water. It would be wise to begin examining national policies for their ability to handle these climate change issues. The CAST report on preparing the U.S. for climate change (CAST, 1992) emphasized the need for climate change-related agricultural research and suggested modifying existing policies to encourage more flexible land use, more prudent use of water resources, and freer trade.

Summary

The three major uncertainties regarding impacts of climate change on agriculture are: (1) the magnitude of regional changes in temperature and precipitation; (2) the magnitude of the beneficial effects of higher CO2 on crop yields; and (3) the ability of farmers to adapt to climate change. Current assessments suggest that, in all three categories, developed nations will frequently be at an advantage compared to most developing nations.

With regard to climate, many developed nations are in mid- to high-latitude locations, where warmer temperatures may improve crop yields by extending the growing season. In contrast, many developing nations are in subtropical and tropical areas, where global warming may lead to excessively high temperatures and reduce yields.

Many crop models account for the CO2 effect by globally increasing yields of C3 crops by 20 - 35%, which assumes a near optimum growth environment. Field conditions are seldom optimum, but farmers with access to adequate water, fertilizer, and other inputs will likely gain more from a CO2 doubling than farmers who do not have these resources. Temperatures may become too hot, or be too low despite global warming, for beneficial effects of CO2 in some areas. It should also be noted that those farmers producing C4 crops, such as maize, sorghum, millet and sugarcane, will see very little benefit from higher CO2, and at the same time their crops will face increased competition from C3 weeds.

Farmers in developed nations will have an obvious advantage in adapting to climate change because of a strong agricultural research base and capital available for farm inputs and infrastructure investments. Food security in these countries may not be directly threatened, and overall productivity may even increase, if global warming is moderate and the frequency of severe weather events does not increase However, adapting to climate change will be costly. Even within countries that benefit at the national level, climate change is likely to have negative economic and environmental impacts in some areas as production zones shift.

Studies integrating climate, crop, and food trade models suggest that a moderate climate change may have only a small impact on world food production because reduced yields in some parts of the globe are offset by increased yields in others. Despite this, severe food shortages are likely to occur in some developing nations because of trade and local climate and resource constraints. This will have political consequences at the global level. Climate change will likely lead to an increase in world hunger unless population growth rates in developing nations are much smaller than currently projected, and farmers obtain adequate assistance. Adapting to climate change with minimal economic, social, and political upheaval will require a coordinated international effort to deal with the many serious consequences of climate change on agriculture.

References

Adams, RM, C Rosenzweig, RM Peart, JT Ritchie, BA McCarl, JD Glyer, RB Curry, JW Jones, KJ Boote, LH Allen, Jr. 1990. Global climate change and US agriculture. Nature 345:219-224.

Council for Agricultural Science and Technology (CAST). 1992. Preparing U.S. Agriculture for Global Climate Change. Task Force Report No. 119. CAST, Ames, Iowa. 96 pp..

Gates, DM. 1993. Climate Change and Its Biological Consequences. Sinauer Assoc. Inc., Sunderland, Massachusetts, 280 pp..

Kaiser, HM, SJ Riha, DS Wilks, DG Rossiter, R Sampath. 1993. A farm-level analysis of economic and agronomic impacts of gradual climate warming. Amer. J. Agr. Econ. 75:387-398.

Kimball, BA. 1983. Carbon dioxide and crop yield: An assemblage and analysis of 430 prior observations. Agron. J. 75:779-787.

Melillo JM, TV Callaghan, FI Woodward, E Salati, SK Sinha. 1990. Effects on ecosystems. In: Climate Change: The IPCC Assessment, JT Houghton, GJ Jenkins and JJ Ephraums (eds.). Cambridge Univ, Press, Cambridge. pp 283-310.

Parry ML. 1990. Climate Change and World Agriculture. Earthscan Ltd. London. 157 pp..

Rosenzweig, C and ML Parry, 1993. Potential impacts of climate change on world food supply: A summary of a recent international study. In: Agricultural Dimensions of Global Climate Change., HM Kaiser and TE Drennen (eds.). St, Lucie Press, Delray Beach, Florida. pp. 87-116.

Rosenzweig C and ML Parry. 1994. Potential impact of climate change on world food supply. Nature 367:133-138,

Wolfe, DW 1994. Physiological and growth responses to atmospheric carbon dioxide. In: Handbook of Plant and Crop Physiology. M Pessarakli (ed.) Marcel Dekker, Inc., New York. pp. 223-242.

Wolfe, DW and JD Erickson. 1993. Carbon dioxide effects on plants: uncertainties and implications for modeling crop response to climate change. In: Agricultural Dimensions of Global Climate Change., HM Kaiser and TE Drennen (eds.). St. Lucie Press, Delray Beach, FL, pp. 153-178.

Go To Health Impacts