Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Agriculture and a Climate 6 Changed by More Carbon Dioxide PauIE. Waggoner 6.l INTRODUCTION The crops that feed us stand outdoors in the wind, rain, and frost. Except for l7 pounds of fish in the l400 pounds each American eats yearly, all the food for us and the feed for our animals, too, grows on a third of a billion acres of cropland and vast rangelands and pas- tures, exposed to the annual lottery of the weather. Although about 50 million acres of American crops are protected from drought by irriga- tion, even these depend on precipitation in the long run, and sheltering crops in greenhouses from temperature extremes is too expensive for staples. Thus, decade after decade grocery prices for all and hunger of the poor puncture the arrogance of our technology and remind us that a few chance degrees of warmth or drops of rain, properly timed, protect us. If it were not so, "Then all the concern about future climate that has been so widespread in recent years would be much ado about nothing" (McQuigg, l979). 6.l.l Concentrating on a Critical, Susceptible, and Exemplary Subject Agriculture is broad. It is the science and art of the production of plants and animals useful to man and in varying degrees their prepara- tion for man's use and their disposal by marketing. This chapter on agriculture could, therefore, encompass forests and fisheries as well as farmers, hunters, and grocers as well as farmers, logs, and hogs Several individuals have assisted the author. Clarence Sakamoto, Norton Strommen, and Tom Hodges assembled the data and used the regression and simulator models to predict changes in yield. Herbert Enoch, Gary Heichel, Robert Loomis, Israel Zelitch, and James Tavares contributed to the section on the effects of CO2 on photosynthesis and plant growth. Marvin Jensen and Glenn Burton advised on water and breeding. The author gratefully acknowledges these contributions, but it is the author alone who takes responsibility for any errors or omissions. 383
384 as well as crops. It could encompass the Ukraine and the Humboldt current, Iowa and the Hoboken docks, California and the Amazonian forest. Although weather surely affects all of these, their survey in a single chapter would be incomplete or superficial, and we have concentrated on a critical, susceptible, and exemplary part: American crop production. Crops are critical because without their photosynthetic conversion of solar to food energy there would be neither bread nor meat. American crops are critical to Americans because they feed us and bring in $40 billion of our foreign exchange (USDA, l982) . And, they are critical to others: for example, in l979 the United States provided 42% of the wheat and l9% of the rice exported by the nations of the world, and fully 43% of the world's corn crop is American (USDA, l982). The susceptibility of crops, rooted in place and exposed outdoors, makes them biologic indicators of weather. American crops growing from 35 to 49 degrees latitude are within the zone that meteorologists predict will experience a change in the weather as CO2 increases, and thus these critical and susceptible crops are also exemplary. Often we shall concentrate further, examining wheat, corn, and soybeans, which outdistance in value any other American crop. 6.l.2 Agriculture and Past Changes in the Weather Later paragraphs will show technical calculations and projections of changes in American crops matching projections of the weather. Some were derived from historical statistics. Changes in the weather cause changes in agriculture that are too complex to be distilled into a few statistics, however, and at this point a background of real life stories is painted. History is, after all, the laboratory notebook where the results of experiments performed by Nature herself are recorded. Nature has not, of course, actually experimented less than l00 ppm with CO2 within the human era. Nevertheless, her experiments with temperature, rain, and snow show in general how farmers are affected by atmospheric change, and then C02 may bring changes in temperature and water themselves. When rain fell abundantly on Italy from 450 to 250 B.C., intensive agriculture was fruitful and the Roman Republic was vigorous. From l00 B.C. to A.D. 50 rain again fell abundantly and Roman civilization was high. After A.D. 80 rain was light, vines and olives replaced cereal, and the Roman Empire declined and fell (Brooks, l970). This demon- strated the adaptation of agriculture by changing cropsâbut the new crop did not sustain the Empire. The years l30l-l350 experienced a change in climate, and Brooks (l970) calculated a maximum of raininess for the half century. Tuchman (l978) wrote that the medieval people were unaware that ...owing to the climatic change, communication with Greenland was gradually lost, that the Norse settlements there were being extinguished, that cultivation of grain was disappearing from Iceland and being severely reduced in Scandinavia. But they
385 could feel the cold, and mark with fear its result: a shorter growing season. This meant disaster, for population increase in the last century had already reached a delicate balance with agricul- tural techniques. Given the tools and methods of the time, the clearing of productive land had already been pushed to its limits. Without adequate irrigation and fertilizers, crop yield could not be raised nor poor soils be made productive. Commerce was not equipped to transport grain in bulk from surplus-producing areas except by water. Inland towns and cities lived on local resources, and when these dwindled, the inhabitants starved. In l3l5, after rains so incessant that they were compared to the Biblical flood, crops failed all over Europe, and famine, the dark horseman of the Apocalypse, became familiar to all. The previous rise in population had already exceeded agricul- tural production, leaving people undernourished and more vulnerable to hunger and disease. Reports spread of people eating their own children, of the poor in Poland feeding on hanged bodies taken down from the gibbet. In this ancient experiment were demonstrations that colder as well as warmer weather can damage; also the length of the season as well as the mean temperature is critical, transportation can alleviate hunger, and the impact of changed climate is sharp at the poleward margin of farming. In l886-l893 an American experiment was performed in the Middle Border, and the Wayne Township (Kansas) Farmer's Club recorded the results (Maiin, l936). Problems were met with full force because movement into the region was swift during the most favorable weather of 30 years. After the farmers were settled, however, "Out of 47l acres of fall wheat there is not wheat enough to cover l5 acres. All winter killed." Although flooding was sometimes a problem, drought was the major force. In l886 Turkey wheat was not favored because millers did not like it, but 4 years of hot summers and cold winters made the hardy Turkey wheat the favorite. Although in l885 the acreage of corn per farm was four times that of wheat, by l905 there were more than 3 acres of wheat for every acre of corn. The predominance of grazing land over cropland ran in cycles with the weather. A final adaptation was flight: in l895 a traveler across the County reported that it was practically deserted. Enough people were left, however, for the subject of "Rainfall and the Populist Party in Nebraska" (Barnhart, l925). The Dust Bowl of the l930s was a natural experiment with results dramatized in John Steinbeck's Grapes of Wrath (l939), and drifting soil is vividly remembered. The catastrophe was not only wilting of crops: pests encouraged by the drought amplified the damage (National Research Council, l976).
386 The red spores of wheat rust were blown over the Wheat Belt and infected wheat; rust fungus erupted from stems and leaves. The greatest losses of wheat during l92l-l950 were in the l930s: 3.4 million metric tons in l935 and l937 were much greater than the runner-up, l million metric tons in l923. Weeds invaded. The prickly pear cactus headed east, invading l.6 million hectares in western Kansas. More young jackrabbits survived in drier weather, and productive rangeland was turned into pastures of prickly pear with a plague of jackrabbits. Grasshopper plagues are generally associated with drought. Like a chapter in Exodus, the report of l936 was of a loss of $l06 million, more than the total gross income from all farm products in Arizona, Nevada, New Mexico, Utah, and Wyoming combined. The hoppers made such a clean sweep in South Dakota that jackrabbits, faced with starvation, escaped into Nebraska (Schlebecker, l953). Finally, an experiment in direct effect on cattle is added. A cattle boom began in Dakota in l883, The Bad Lands Cow Boy claimed, "We have never heard of a solitary head ever having died in the Bad Lands because of exposure," and Theodore Roosevelt invested. In l886-l887 storm piled on storm, children were lost and froze within a hundred yards of home, and cattle, desperate for shelter, smashed their heads through ranch house windows. The average loss of cattle was 75%, and Roosevelt rode for 3 days the following summer without seeing a live steer. Extreme weather rather than the average affected farming. A change of weather and the consequent boom to bust took only 3 or 4 years (McCullough, l98l). Nature's experiments demonstrated the following: Farmers fit husbandry and crops to the weather. Swift change disrupts. Colder as well as warmer and wetter as well as drier can damage. Pests amplify effects of bad weather. â¢ The very soil can be changed by weather. Even warm-blooded animals are affected by weather. - Occasional extremes destroy agriculture. Impact of changed weather is sharp in marginal climates. Migrations and political upheaval are blown by bad weather. Empires rise as well as fall as yields wax and wane in changing climates. Against this background we shall now examine the range of change meteorologists set before us and then calculate or speculate from science on the changes in crops that would follow. 6.l.3 The Range of Change in the Atmosphere Carbon Dioxide. The CO2 concentration of the atmosphere has risen from below 300 parts per million by volume (ppmv) in the l800s to about 340 ppmv at present. It is estimated that by the year 2025 it will
387 likely reach 425 ppmv, and by 2065 it will probably pass 600 ppmv (Nordhaus and Yohe, this volume, Chapter 2, Section 2.l). Nordhaus and Yohe's upper estimate for CO2 concentration in the year 2000 is 400 ppmv (Chapter 2, Figure 2.23). We shall see that CO2 has a direct effect on crops. Temperature. Available climate models indicate that a doubling of the CO2 content could raise the global annual average surface temperature by 3Â°C (this volume, Chapter 4). If we assume the rapid rise in atmospheric concentrations to 400 ppmv in A.D. 2000, the temperature rise by A.D. 2000 would be about lÂ°C. In polar latitudes a doubling of the atmospheric CO2 concentration would cause a 5 to l0Â°C warming (Mitchell, l977). The polar and higher latitudes are sensitive because of summer changes in the albedo of regions normally covered by snow and ice throughout the year. For middle latitudes, Manabe and Stouffer (l980) calculated a lesser warming: 3Â°C at the U.S.-Canadian border after a doubling of CO2. Length of Growing Season. Kellogg (l977) pointed out a simple relation between summertime mean temperature and the length of the growing season at middle and high latitudes: A lÂ°C change in mean temperature for the summer corresponds approximately to a l0-day change in the length. We must beware that the predicted annual average temperatures include winter, and conceivable changes in daily amplitudes or vari- ability (Neild, l979) could modify the simple relation. Using this l0-day rule of thumb, however, one can reasonably consider the effect on crops of a lengthening of the growing season by l0 days in the northern United States where a lÂ°C increase in average annual temper- ature has been predicted by A.D. 2000. Precipitation and Moisture. Manabe and his colleagues (l98l) have projected the changes in soil moisture that will accompany the predicted rise in temperature with three models of increasing geographic detail. They estimated the change from present CO2 concentrations to four times as much. The three models all predicted drier summers at middle and high latitudes. Snow would melt earlier at high latitudes, causing an earlier transition from spring to summer and less rain. The earlier onset of summer would evaporate and transport more moisture from the soil. Conclusion. Looking toward a horizon near A.D. 2000, the agricultural- ist may be uncertain how the climate of a precise place will change, but he can reasonably consider how crop production would be changed by an increase to about 400 ppmv, a mean warming of about lÂ°C in the northern United States with a growing season about l0 days longer, and more frequent drought in the United States caused by somewhat less rain and slightly more evaporation.
388 6.2 EFFECTS OF C02 ON PHOTOSYNTHESIS AND PLANT GROWTH Carbon dioxide is a major substrate for photosynthesis and, therefore, can directly affect plant growth if C02 is limiting. The current 340 ppmv appears to be limiting (Figure 6.l) and, therefore, a rise in atmospheric CO2 levels should increase photosynthesis. However, most effects of CO2 on photosynthesis and plant growth have been studied and measured during short periods where other factors such as light, water, temperature, and nutrients are optimal. In addition, growth habits and adaptations to different environments might mitigate or alter the effects of changing C02 concentration. Yield is the important integration of physiology, and it is difficult to predict the changes in yield that might follow a rise to 400 ppm of C02. This section summarizes the current understanding of the direct effects of C02 on plant growth and the factors that must be considered in projecting these effects to yields. More discussion of direct effects can be found in the proceedings of the International Conference on Rising Atmospheric Carbon Dioxide and Plant Productivity, Athens, Georgia, May l982 (Lemon, l983). In addition to showing the direct effect of CO2 on plants, the curves of Figure 6.l are critical in calculating the rise in CO,, itself. The percent change in photosynthesis per percent change in CO2 in the air, which is typically covered by the term "beta" in carbon cycle models (c.f. Woodwell, this volume, Section 3.3.3), must be about 0.25 to account for the rise caused by burning fossil fuel since the Industrial Revolution (Bjorkstrom, l979). The beta of the illuminated wheat leaf in 400 ppmv C02 (Figure 6.l) is 0.8 but of the corn leaf is 0.l. These botanical estimates from the laboratory certainly do not conflict with the estimate of 0.25 from the global fuel consumption and atmospheric C02. 6.2.l Photosynthesis About 90% of the dry weight of plants derives from the reduction of CO2 to carbohydrates by photosynthesis. In single leaves in bright light, net photosynthesis increases with CO2 above the current atmospheric level of 340 ppmv, and this is confirmed in whole plants by greater crop yields (see Section 6.2.6). On the basis of photosynthetic properties plants are classified into C3 plants, C4 plants, and crassulacean acid metabolism (CAM). Although photosynthesis occurs in all plants by the C3 pathway, the C4 and CAM plants have specialized steps to sequester C02 into the leaf. In C4 plants C02 is incorporated into C4-dicarboxylic acids that can be transported within the leaf to sites where C02 is released for photosyn- thesis. In specialized desert plants of the CAM type the pores in the leaves (stomates) open at night to collect CO2 into organic acids that later regenerate the C02 inside the leaf for photosynthesis during daylight. This allows CAM plants to keep their stomates closed during the heat of daylight and thereby save water.
389 100 7_ g 50 Maize 200 400 600 C02 (ppmv) 800 FIGURE 6.l Typical photosynthesis response of plants to CO2. Net photosynthesis of wheat is about 70 mg of CO2 dm"2 ITl compared with maize (about 55 mg of CO2 dm"2 h-l) for equivalent light intensity (0.4 cal cm"2 min-l). Maize is saturated at a lower CO2 concentration (~450 ppm) than wheat (~850 ppm). CO2 in ppmv is percent CO2 x l04. (Adapted from Akita and Moss, l973.) The same enzyme that catalyzes the first step in the reduction of CO2 to carbohydrates can also oxidize the first product. This oxidation occurs in light also but uses O2 to oxidize carbon back to CO2. It is called photorespiration, and its rate is determined by the ratio of CO2 to O2 within the leaf. Photorespiration rates are high in C3 plants. By increasing CO2 levels from 340 to 400 ppmv, CO2 uptake is enhanced about 20% in C3 (high-photorespiration) species and about 7% in C4 (low-photorespiration) species (Hesketh, l963; Akita and Moss, l973). Thus, increasing CO2 directly benefits C3 species more than C4 species. Rising CO2 can be expected to alter leaf carbon metabolism and thereby affect the rate and duration of photosynthesis and the fate and par- titioning of the photosynthate. 6.2.l.l Rate of Photosynthesis The faster CO2 fixation per leaf area at higher CO2 may be explained by the kinetic properties of ribulose bisphosphate carboxylase, the primary CO2-fixing enzyme in the chloroplast, and to some extent by the con-
390 sequently slower photorespiration in C3 species. The carboxylase is a large molecule (mol wt = 560,000) that comprises about half the total chloroplast protein and about a quarter of all the protein in the leaf. Half-maximal rates of CO2 fixation with the isolated enzyme are obtained at about 600 ppmv of CO2 (Jensen and Bahr, l977), which is not greatly difficult from the curvilinear increase in net photo- synthesis of leaves in rising CO2 (Figure 6.l). Neither the photo- chemistry and electron transport nor the rate of regeneration of the CO2 acceptor by the photosynthetic carbon reduction cycle appear to limit photosynthesis in the range of CO2 under consideration. The release of photorespiratory CO2 is faster at higher O2 concen- trations, while increasing CO2 levels inhibit photorespiration (Zelitch, l97l). The competition between C02 and O2 for photosynthesis and photo- respiration is expressed in C-j species as an inhibition of photosyn- thesis by 02, or as an "oxygen stress" (Zelitch, l982). A large propor- tion of the inhibition of photosynthesis by O2 is attributable to photorespiration, because losses of photorespiratory C02 can be as high as 50% of net photosynthesis in C3 plants (Zelitch, l982). Photosyn- thesis in tobacco leaves was inhibited 35% at 340 ppmv of CO2 at 2l% of O2 compared with 3% of 02 and was inhibited 3l% at 400 ppmv of CO2 (R. B. Peterson, Connecticut Agricultural Experiment Station, New Haven, Connecticut, personal communication, l982) . Thus, the benefit of regulated photorespiration appears less than the advantage of faster carboxylation when CO2 is raised from 340 to 400 ppmv. 6.2.l.2 Duration of Photosynthesis Prolonged and faster photosynthesis caused by increased CO2 in bright light produces more sucrose, sometimes increasing starch accumulation in the chloroplasts, and excessive starch can deform chloroplasts and decrease photosynthesis (Guinn and Mauney, l980). Besides increasing sucrose, faster photosynthesis may change the levels of phosphorylated compounds and decrease orthophosphate levels in chloroplasts, which feed back to inhibit photosynthesis and decrease CO2 assimilation (Walker, l976). Such negative feedback might be bred against to obtain the full benefits of increased CO2 on yield. Yield depends on an adequate storage or sink to accept the products of photosynthesis. If the sink is inadequate, feedback will decrease photosynthesis. Sink capacity and yield tend to increase in parallel until they reach the limit set by the photosynthetic capacity (Evans, l975). 6.2.l.3 Fate and Partitioning of Photosynthate The products of photosynthesis and the efficiency of their translocation to the sites of conversion to starch, protein, and lipids may affect photosynthesis itself as well as the accumulation of carbon in the storage organ. Higher CO2 decreases photorespiration and thus indirectly will affect nitrogen metabolism, since ammonia turns over
39l rapidly in leaves during photorespiration, and the balance of amino acids available for storage will change (Lawyer et al., l98l). Although less lipids might be made in leaves, more may be produced in storage organs that synthesize lipids from translocated sucrose. These specula- tions are based on sound biochemistry, but data are lacking. 6.2.2 Drought CO2 directly affects the water in plants because CO2 affects the pores or stomata through which water is transpired. The epidermis of leaves is generally perforated by countless microscopic stomata, where the size of the opening is regulated by the shrinking and swelling of two guard cells that border it. The C02 for photosynthesis is acquired through these pores, and since the pores open into the moist interior of the leaves, water is transpired. Most stomata close in the dark, conserving water when photosynthesis stops. In fact, mere narrowing of the pores conserves water, even in a crop with several acres of leaf surface area per acre of land (Waggoner et al., l964). By regulating the size of stomata, plants can seek a balance between the necessary uptake of CO2 and the stresses caused by excessive loss of water. We noted earlier the unique method used by CAM plants to acquire CO2 under arid conditions. In bright light, maize stomata narrow when CO2 concentration increases from 300 to 600 ppmv, and transpiration decreases about l5% from the potted plants. Transpiration from pots of wheat decreased only about 5%, however, and the difference between wheat and corn is assumed typical of the difference between C3 and C4 plants (Akita and Moss, l973). In crops in the field, Baker (l962, l965) observed that transpiration of corn and cotton in bright light decreased by 20 and 35% between 300 and 600 ppmv. The loss of water from the soil includes evaporation from the soil surface itself, especially when it is moist and unshaded by foliage, and from wet foliage as well as transpiration through stomata. Thus the loss of water from a field, month in and month out, will be affected less by C02 than by the percentages of 5 to 35% given above. Although we know of no experiments showing that crops can tolerate drought in C02-rich air, there is some foundation to expect an increase to 400 ppmv of C02 would decrease transpiration from a crop with abundant foliage by the order of 5%, somewhat more in C4 and somewhat less in C3 crops. Since reduced transpiration would deplete soil water somewhat more slowly, drought would logically be somewhat less frequent. 6.2.3 Nutrients The demand for nitrogen and mineral nutrients in plant growth is tied to photosynthesis. As photosynthesis increases with increasing C02, the carbohydrate available for plant growth will increase and, in turn, impose demands for increased fertilizer or available soil nutrients. Since plant biomass typically has a minimum nitrogen content of between
392 2 and 3% of dry matter and a phosphorus content near 0.2 to 0.3%, nutri- ents may be needed to realize any increase in production. C02 may improve the availability of nutrients: Photosynthates are utilized as energy by symbiotic nitrogen-fixing organisms. The nitrogen in plants with such symbionts associated with their roots may increase. For the same reason more photosynthates might increase uptake of N, P, and K by roots in their association with bacteria and fungi. And finally, increased photosynthesis and growth could enlarge the pool of decomposing soil organic matter that can serve as a reservoir of soil nutrients. 6.2.3.l Nitrogen Metabolism Nitrogen fixation consumes much photosynthate for its energy. Nitrogen- fixing symbionts such as Rhizobium and Frankia, therefore, are greatly affected by the photosynthate available in the plant. While quantita- tive data are limited, experiments with CO-enriched plants show increased N2 fixation. In field experiments, Hardy and Havelka (l977) showed more N2 fixed in soybeans in air enriched with C02 to 800-l200 ppmv, and the total kilograms of N2 fixed per hectare over the growing season was much higher with C02-enrichment. In terms of harvestable product, Rogers et al. (l980) found a 28% increase in weight of seeds harvested per soybean plant at 520 ppmv of CO2 in comparison with 340 ppmv. Of all the essential elements for plant growth it is usually nitrogen that is limiting. Thus the increased photosynthesis in a CO2-enriched atmosphere will increase demand for nitrogen. There also might be a shift to plants capable of N2 fixation and to plants that have a beneficial association with free-living, N2-fixing microbes. The increased nitrogen in these plants would permit increased growth and, in turn, increased photosynthesis, more photosynthate, and even more N2 could be fixed until some other factor became limiting. 126.96.36.199 Organic Matter and Rhizosphere Association Increased atmospheric C02 will not likely affect the rhizosphere. The respiration of roots and soil microbes maintain C02 concentra- tions in the air spaces in soil that are l0 to 50 times higher than in the atmosphere. Doubling atmospheric CO2 would likely have little direct effect on roots and soil microbes. If CO2 increased plant growth it could increase the plant remains incorporated as soil organic matter and influence the cycling of min- erals in the soil and other soil properties. Soil organic matter is composed of two major factions that occur in roughly equal amounts. One factor is rapidly turning over and is composed of readily metabo- lized organic molecules such as cellulose; the other is composed of slowly accumulating, stabilized aggregates of phenols, other aromatic molecules, and inorganic particles with turnover times of at least l00 years (Van Veen and Paul, l98l). Although increases in degradable
393 biomass from green manuring generally raise the activity of soil microbes and change soil organic matter only slightly, CO2 will be recycled to the atmosphere faster in response to the increase of degradable organic matter. With increased substrates in the form of root exudates or degradable soil organic matter there may be increased nitrogen fixation by bacteria, both free-living and symbiotic, and increased mycorrhiza. Many minerals required for growth, such as P, Zn, and Cu, are unavailable to roots because they are predominantly in immobile and insoluble forms in the soil. Micorrhizal fungi may increase their availability to host plants by penetrating more soil. Thus, increased photosynthesis and degradable biomass are likely to increase soil nitrogen levels, perhaps by 5 to l0%, and may slightly decrease the phosphorus and soil nutrients not tied up in living biomass (Lemon, l983). 6.2.4 Phenology The development of new organs is distinct from their mere enlargement or growth, and one can reasonably ask if this development is affected by CO2 in the air. Although one plant, cucumber, flowered earlier and the fruit was ready for market two weeks earlier when the air was enriched with CO2, another plant, pepper, proceeded at the same rate (Enoch et al., l970). CO2 did not change the period from pruning to harvest of roses (Zieslin et al., l972), and a tenfold increase in CO2 scarcely shortened the time from flowering to ripe strawberries (Enoch et al., l976). Thus changed timing of flowering and other stages in the life of crops is not likely to be an important consequence of the rise in C02 that we are considering. 6.2.5 Weeds Environment affects crops, their pests, and the relation between them. Among the insect, disease, and weed pests of plants, it is the weeds fueled by their own photosynthesis that can be directly affected by changing atmospheric C02. We shall examine the effects of insects and disease on crops later in this chapter in the context of changes in temperature and moisture. Annually weeds exact a toll of some $l8 billion in the United States by competing with crops for light, water, and fertilizer (Chandler, l98l). Because C3 and C4 plants respond differently to changes in atmospheric C02 concentrations, the generalities of weeds versus crops have been examined in those terms. Will the competition tip toward the weed if it is C3 and benefits from an increasing CO2 concentration while the crop is C4 and responds less to increasing CO2? Twelve of the l5 crops that feed the world are C3 plants (Harlan, l975), whereas l4 of the l8 most damaging weeds are C4 (Patterson,
394 l982). Although that seems good fortune, l9 of the 38 major weeds of American maize, a C4 plant, are C3 plants (USDA, l972). The generality of Cj gaining relative to C^ plants in rising CO2 has been demonstrated by crops and weeds. The C3 weed velvet leaf (Abutilon theophrasti) increased its growth more than maize when CO2 was increased. On the other hand, the C4 weed, itch grass (Rottboellia exaltata), gained less than the C3 crop, soybean (Glysine max) (Patterson and Flint, l980). In addition to a gradual increase of a few percent in growth, C02 can conceivably remove a limitation to the spread of a weed. Thus, okra, which is a crop becoming a weed, can grow at lower temperature and presumably at higher latitudes if CO2 is enriched (Sionit et al., l98lb) . Limitations of fertilizer, water, or light might of course limit the realization of an advantage of C02 to a weed or crop. In fact, how- ever, limited nutrients and water have failed to nullify the benefits of C02, including those to the height and leaf area that will affect competition for light (Patterson and Flint, l982). Thus increases in atmospheric C02 may affect the competition of weeds and crops, some- times to the advantage of the crop and sometimes to the weed. 6.2.6 Direct Effects of CO? on Yield The integration and the practical outcome of all the effects enumerated above is yield. Calculating the direct effect of increased C02 on yield is chancy because of lack of experimental data. Few food, feed, or fiber crops have been grown at elevated CO2 from sowing to harvest. Most experiments have been brief, with emphasis on a specific stage of growth, and have been conducted with flowers and ornamentals rather than crops. Kimball (l982) recently summarized the results of 70 C02 enrichment experiments conducted during the past 64 years. In Table 6.l the results of several experiments are presented for the entire life cycle of crops. All experiments were in growth chambers and greenhouses; no results from fields were available. In all experiments plants were grown in optimum environments without pests and with abundant water and nutri- ents. We cannot claim that light was optimum because plants seldom enjoy this status when grown in chambers and greenhouses. Although all experiments were performed in equable environments, all experiments were, nevertheless, dissimilar. Some experiments included stressful conditions as treatments. As these stresses might reflect a situation encountered in the field, their effect on response of yield to CO2 was important to our evaluation. First beta, here taken as the percent change in yield per percent change in C02, is examined. The range is a 0.l to 0.9% increase in dry weight per percent increase in CO2. As in the estimation of beta from the simple response to C02 shown in Figure 6.l, the global estimate of 0.25 for beta (see Woodwell, this volume, Chapter 3, Section 3.3) is not denied by these additional botanical estimates. In addition, there is evidence in Table 6.l that drought or lack of
395 TABLE 6.l Changes in Yields of Crops in Optimum and Stressful Environments Anticipated from Atmospheric Enrichment to 400 ppmv of CO2 Crop Change in Yield Component (%) Harvested Yield Increment/ Yield CO2 Change by Increment Enrichment (%/ppmv (%/60 ppmv of CO2) of C02) Reference Optimum Environments Barley 0.9*. Corn 0.28 Cotton 0.6 Soybean 0.43- Wheat 0.4 Wheat 0.3 Wheat 0.6 Stressful Environments Grain 0.l8 Young shoots 0.03 Lint 0.34 Grain 0.04 Grain Grain Grain 0.l3 0.07 0.l3 ll Gifford et al. (l973) l.9 Wong (l979) 20 Mauney et al. (l978) 2 Hardman and Brun (l97l) 8 Gifford (l979) 4 Sionit et al. (l980) 8 Sionit et al. (l98la) Corn 0.28 Young shoots 0.03 l.9 Wong (l979) (l/3 normal N) Wheat 0.6 Grain 0.44 26 Gifford (l979) (water limited) Wheat 0.5 Grain 0.l0 6 Sionit et al. (l980) (l H2O stress cycle) Wheat 0.2 Grain 0.05 3 Sionit et al. (l980) (2 HjO stress cycles) Wheat 0.l Grain 0.02 l Sionit et al. (l98la) (l/8 normal nutrient) ^Calculated from shoots only. fertilizer will not prevent any increase in photosynthesis and a sequestering of more CO2 as its concentration increased. Proceeding to the yield of grain or cotton lint, one sees increases of 0.07 to 0.34% per ppmv increase in C02 in optimum growing environ- ments. There is no clear evidence that this relative change is less in wheat that lacks water, but the wheat deprived of nutrients did respond little to C02. We conclude from the examples in Table 6.l and from the extensive survey of Kimball (l982) that C02-enrichment to 400 ppmv by A.D. 2000 may increase the annual yields of well tended crops by, say, 5%. In comparison, the yield of Illinois corn roughly quadrupled during the past half century, and about three fourths of the change is attributable to improved husbandry (Haigh, l977). Although the evidence of Table 6.l for poor growing conditions is equivocal about quantity, some increase in yieldâeven in poor circumstancesâis indicated.
396 6.3 PREDICTING THE CHANGES IN YIELD THAT WILL FOLLOW A CHANGE TO A WARMER, DRIER CLIMATE Two orderly means are at hand to calculate how the yields of corn, soybeans, and wheat will change if rising CO2 makes the climate warmer and drier. In one method, the statistical regression model, history is distilled to obtain the change in yield for a specified change in weather, for example, the decrease in quintals of wheat/ hectare in Kansas after a l0% decrease in March precipitation. In the other method, the physiology of wheat and the physics of evaporation are assembled and used to compose a computer program or "simulator" of wheat to calculate the change in wheat yields per change in environment. We shall employ both these methods, comparing and verifying them as well as making predictions. More must be said of the climatic change to be caused by an increas- ing concentration of atmospheric C02 than simply "lÂ°C warmer and a bit drier." The direct effects of C02 on plant growth have already been dealt with in the preceding section and are not incorporated here. To approximate the CO2-induced weather change for the year 2000, we shall employ the actual weather data for a location, which includes the natural correlations between temperature and rain, and then increase the temperature by lÂ°C and subtract l0% of the precipitation. The reader must remember the vagueness of predictions of changes and realize the lÂ°C and l0% drier is merely a rational example and not a prediction. The corresponding change in the length of the growing season will not affect the historical models unless spring and fall temperatures are employed and are proxies for frost; the growing season will be explicitly changed in the simulations. The change in evapora- tions with lÂ°C warming will implicitly enter the historical models and be explicitly calculated in the simulations from the increased warmth. How shall the results be presented? The historical models provide coefficients of quintals/hectare/millimeter and so forth that we can examine. The simulators allow us to calculate the frequency distribu- tions, year by year, of yields in the present climate and the distribu- tions, year by year, if the climate, on the average, warms by lÂ°C and lacks l0% of the precipitation. This is a more informative result than a single normal or average, which so rarely occurs. 6.3.l History For decades the yield of a "crop reporting district" has been estimated after each growing season. A crop reporting district often includes several counties. For example, Iowa is composed of nine crop reporting districts. The Cooperative Climatological Service of the National Weather Service has been recording temperature and precipitation daily at about a dozen stations in each district. Using these data, Thompson (l969) estimated the effect of weather by correlating the yields with the weather.
397 Crop yield, temperature, and precipitation for a district are used in statistical regression models. The model for a particular crop in a particular area is generally expressed as Yi - a + b1ti = b2X2i + + bnXni, where Y^ = estimated yield in the ith year; a = intercept; t^ = surrogate for technology in the ith year; bl Â« coefficient representing the effect of technology in quintals/hectare/year; b2 to bn = coefficients representing the effect in quintals/hectare/unit change in the weather; X2i to xni = weather variables such as precipitation, temperature, potential evapotranspiration (PET), and evapotranspiration (ET) in the ith year. Thus, by multiple linear regression the effect of weather, factor by factor, on the yields of crops is distilled from a history of weather and yields, accumulated for decades by faithful observers. Beginning in the mid-l970s a comprehensive set of estimates of the b coefficients was made by the National Oceanic and Atmospheric Adminis- tration, the U.S. Department of Agriculture, the National Aeronautics and Space Administration, and the University of Missouri for the three major crops in the states of the nation's grain belt. In some cases the temperature and precipitation were combined into variables repre- senting evaporation or periods of excessive heat (Sakamoto, l978). Our results are presented state by state. The results for a state, however, were calculated from regressions for the several districts in a state and then aggregated to the whole state because this provides a better estimate for a state than directly relating state yield to state weather (Sakamoto, l982). For example, yields estimated for the nine districts in Iowa and then aggregated are within about 8% of the actual state yield (LeDuc, l980). The parameters that we shall show for each state are aggregations of the districts of a state. The effect of weather on wheat is seen in Table 6.2. Look at the column for the Red River Valley, which is northern and likely to experience a change in climate from rising CO2. The valley includes eastern North Dakota and western Minnesota. The increase in yield for a lÂ°C warming in April, the planting season, is represented by a b coefficient of 0.2l quintal/ha. On the other hand, a lÂ°C warming in June or July will increase the frequency of hot days over 32Â°C and yield will be decreased by 0.25 and 0.42 quintal/ha for each hot day added to June and July. In the Red River Valley, spring wheat is heading in June and July, and hot spells during heading decrease yield. In Kansas, heading occurs in winter wheat around May; therefore hot spells in Kansas in May decrease yields there. South Dakota shows the effect of weather in the transition region between winter and spring wheat where spring wheat is relegated to unfavorable sites where
398 TABLE 6.2 The Effect of Weather on Yields Wheat (WW) and Spring Wheat (SW)3. (in quintals/ha) of Winter Crop Region, State Red River North South Valley Dakota Dakota Nebraska Kansas Oklahoma (SW) (SW) (SW) (WW) (WW) (WW) A. Variable Yield, Average l978-l980 l8.2 Temperature, Â°C Jan to Feb Apr 0.2l May July Temperature, No. of days above 32Â°C May June -0.25 July -0.42 Precipitation, mm Sept to Nov Aug to Nov 0.0l Sept to Dec Jan to Peb â Mar May June Combined Variables Mar Free minus May Free minus Apr Prec/PET0- (Apr Prec/PET)2 B. Calculated Estimated Changes l4.9 l2.0 2l.3 -0.24 -0.74 -0.69 -0.42 -0.2l 0.02 0.02 0.02 -0.0l -0.29 -0.22 0.0l 0.02 -0.02 -0.04 0.7l -0.322 â 2l.3 l9.7 -0.30 -0.l6 0.02 0.0l 0.0l 0.06 0.03 -0.02 -0.0l -0.02 -0.0l Yield, quintals/ha -l.32 -l.77 -l.36 -l.04 -l.04 -0.37 Change from l978-l980 -7% -l2% -ll% -5% -5% -2% average Â£A. The effects are given as b coefficients in quintals/ha/unit of variable, i.e., mm of precipitation, Â°C of temperature, or fraction of a ratio (after Sakamoto, l978). B. Estimates of the change in yield with a lÂ°C increase in temperature and a l0% decrease in precipitation from the historic average temperature and precipitation recorded for the regions. Â£PET, potential evapotranspiration in millimeters, a measure of the demand for water estimated from temperature. âThe variable Apr Prec/PET is calculated as the deviation from normal. This combined variable and its square produce a curvilinear response with both too little and too much rain relative to the demand, PET. Either extreme in rainfall is harmful.
399 the higher yielding winter wheat will not prosper. Hence, in South Dakota higher temperatures in July decrease wheat yields 0.69 quintal/ha for each degree centigrade of warming. The effect of precipitation on yield also can be seen in Table 6.2. In Kansas, a decrease of l mm in March precipitation causes a decrease in yield with a b coefficient of 0.06 quintal/ha for each millimeter decrease in rain. In addition, precipitation (Prec) can be combined with a measure of the demand for water, potential evapotranspiration (PET), in the combined variable April Prec/PET as is shown in Table 6.2 for South Dakota. A decrease in precipitation would also mean an increase in potential evapotranspiration. Thus a decrease in April precipitation in South Dakota of about l0% would decrease the ratio April Prec/PET about l6%, causing a very small decrease in yield. To appraise the consequences of climatic change caused by rising COo we must, however, examine the consequences of both warmer and drier weather. For example, to obtain an estimate of the change in wheat yield in South Dakota if weather becomes lÂ°C warmer and l0% drier one needs only to multiply the b coefficients in Table 6.2 by the change in the recorded weather variables normal for the region. A lÂ°C increase in temperature might mean two more days in June with temperatures over 32Â°C. A l0% decrease in the September to November precipitation is a decrease of 8 mm of rain for this 3-month time period. Combining a lÂ°C increase in July temperature, two more days in June over 32Â°C, an 8-mm decrease in September to November precipitation and a l6% decrease in April Prec/PET would reduce the average yield by ll% from l4 to l2.53 quintals/ha. The calculation is (+lÂ°C)(-0.69 yield/Â°C) + (2 days)(-0.29 yield/day) + (-8 mm)(0.0l yield/mm) + (-l6% ratio)(0.7l yield/ratio) + (-l6% ratio)(2)(-0.32 yield/ratio). Because the relation between the ratio April Prec/PET and yield is curvilinear, the effect of decreasing it l6% is negligible. In short, a reasonable consequence of a C02 increase and climate change is an ll% decrease in the wheat yield for South Dakota of l.36 quintals/ha. At a southern location, Kansas, the combined changes in temperature and rainfall produce a somewhat smaller relative decrease. Two more very hot days would occur in May; a l0% decrease in the August to November precipitation is a decrease of 22 mm, and the l0% decrease in March and June precipitation would be 4.5 mm and l0 mm, respectively. The outcome of all this is a l.04 quintal/ha loss or about a 5% decrease in yield in comparison with the average yield for the 3 years l978-l980. The effect of weather on corn is seen in Table 6.3. In Illinois a warming of lÂ°C in July decreases yield l.56 quintals/ha or about 2% of the 3-year average yield of 68.8 quintals/ha. July is the time of pol- lination, and at this stage corn is most sensitive to heat. In August, after pollination, the effect of heat is only half that in July. Warmth in October raises yields 0.57 quintal/ha/Â°C because October is frost time. July is a critical time for precipitation as well as heat. A l0% decrease in July rain and correspondingly an increase in potential evapotranspiration would decrease July Prec minus PET by about ll mm and yield by 0.28 quintal/ha. A decrease in precipitation during planting in May, on the other hand, increases yield.
400 TABLE 6.3 The Effect of Weather on Yields (in quintals/ha) of Corn in Three States3- Crop Region, State Iowa Illinois Indiana A. Variable Yield, Average l978-l980 Temperature, Â°C July Aug Oct July to Aug Average Precipitation, mm May Sept to June Sept to June SDFN^ July Combined Variables Apr and May PET2 May Prec/PET July Prec minus PET 72.7 68.8 -l.56 -0.64 0.57 0.0l3 -0.000l -0.l2 0.076 (June ET/ET^ + July ET/ET)/2 â B. Calculated Estimated Change -0.00006 -l.49 0.025 65.3 -2.34 -0.0l7 0.045 2.27 Yield, Change quintals/ha from l978-l980 -2. 36 -l. -3% 72 -2. 80 average -3% -4% ^A. The effects are given as b coefficients in quintals/ha/unit of variable, i.e., mm of precipitation, Â°C of temperature or fraction of a ratio (after Leduc, l980) . B. Estimates of the change in yield with a lÂ°C increase in temperature and a l0% decrease in precipitation from the historic average temperature and precipitation recorded for the regions. J^SDFN, departure from normal precipitation, squared. Â£PET, potential evapotranspiration in millimeters, a measure of the demand for water. dj3T/ET, evapotranspiration divided by average evapotranspiration. To appraise the consequences of a combined warming and drying of Illinois by rising CO2, we decreased the ratios of May Prec/PET by l2%, decreased precipitation by l0%, and raised temperatures lÂ°C. The net effect is a decrease of l.7 quintals/ha, or 3%. The decrease is moderated because less rain is favorable for corn planting in May but unfavorable for grain development and growth during July to September. Likewise, a favorable, warmer October moderates the effect of an unfavorable increase in temperatures during July and August.
40l Thus a steady warming and drying throughout the year can produce a modest effect on yield because warmer and drier is beneficial at planting and harvest, whereas it is harmful at the height of the summer. In Iowa, unlike in Illinois, the warmer spring is unfavorable as shown by the negative coefficient for April and May PET, the decrease in September to June precipitation is unfavorable, and the warmer summer decreases July Prec minus PET. The net is a greater decrease in Iowa, 2.4 quintals/ha, than in Illinois, showing geographical variations in the consequences of the same change in climate. The effect of weather on soybeans is seen in Table 6.4. Yields are normally about 22 quintals/ha in the three states of Iowa, Illinois, and Indiana. In Iowa a wet winter makes a late spring and a high September-to-May precipitation; a decrease of l0% or 47 mm in September- to-May precipitation would increase yield by 0.2 quintal/ha. A warming of lÂ°C in June would increase yield of the warmth-loving soybeans by 0.43 quintal/ha and in September by 0.35 quintal/ha. The effect of summer rain in Iowa is in the ratios of actual evapo- transpiration as a fraction of the long-term average. Assuming that evapotranspiration by the plants is limited by the water available, a l0% decrease in rain would decrease evapotranspiration (ET) by 0.l and yields by l.9 or 0.6 quintals/ha, respectively, if the decrease were in July or September. The effect of changes in both temperature and rainfall in Iowa were calculated for a l0% decrease in precipitation and a decrease of l0% in the ratio of ET to ET with a lÂ°C warming. The net effect is a loss of l.5 quintals/ha or a 7% decrease below l978-l980 average yields. Now we see how history can foretell the outcome of a changed climate. A couple of remarks are in order: "multiple collinearity" can mislead us, and simpler parameters might be obtained. If two climatic factors are correlated, their correlation changes the regression coefficient, say, quintals per hectare per degrees Celsius in the tables. Sometimes the correlation will cause the coefficient to overstate and sometimes understate the effect of the factor, say, tem- perature on yield. It may even produce a nonsensical value. This is the problem of multiple collinearity. Fortunately in Tables 6.2, 6.3, and 6.4 the coefficients generally make good agricultural sense. In fact, the problem of multiple collinearity has been minimized in the tables by combining variables. For example, July precipitation and temperature are correlated, and both affect Iowa corn. The two were combined in a single variable, July Prec minus PET, by calculating PET from temperature and subtracting it from rainfall, making a single factor expressing the harm to corn from drought in July or the benefits from timely rain during tasseling. The invention of combined variables to avoid collinearity problems and the importance of different seasons in different states with different crops produced the plethora of variables in the tables. Likely, repeating the calculations with an eye for simplicity and standardization could reduce the plethora. Nevertheless, history has been distilled into parameters in the tables for an objective forecast of the agricultural consequences of a climate changed by more CO2.
402 TABLE 6.4 The Effect of Weather on Yields (in quintals/ha) of Soybeans in Three States3- Crop Region, State Iowa Illinois Indiana A. Variable Yield, Average l978-l980 23.6 2l.9 22.0 Temperature, Â° May â â 0.l3 June 0.43 â 0.35 Sept 0.35 Precipitation, mm Sept to May -0.0042 -0.003l July â â 0.0l9 July Free SDFN^ â -0.0003 -0.0002 August ~ 0.0l6 â Combined Variables July Free minus PETÂ£ â 0.039 July ET/ET! l9.24 Aug ET/ET â â ll.83 Sept ET/ET 6.06 l.99 B. Calculated Estimated Change Yield, quintals/ha -l.55 -0.82 -l.25 Change from l978-l980 average -7% -4% -6% 5A. The effects are given as b coefficients in quintals/ha/unit of variable, i.e., mm of precipitation, Â°C of temperature or fraction of a ratio (after Motha, l980). B. Estimates of the change in yield with a lÂ°C increase in temperature and a l0% decrease in precipitation from the historic average temperature and precipitation recorded for the regions. bfiDFN, departure from normal precipitation, squared. Â£PET, potential evapotranspiration in millimeters, a measure of water demand. evapotranspiration divided by average evapotranspiration. In the large, the tables and our calculations have shown some geographical variation in the outcome of climatic change. For example, there are substantial differences in the average yield between spring and winter wheat, therefore, a decrease of l.36 quintals/ha in South Dakota is an ll% decrease in yield, whereas a decrease of l.04 quintals/ha is a 5% drop in yield for Kansas. Also, the season of sensitivity to weather is earlier in Kansas than northward in South Dakota. Although we cannot appraise the consequences, it is well known that the relative variability of precipitation and hence variability of yield increases as the amount decreases.
403 The effect of a change in weather is a net of several biological impacts and hence b coefficients. Thus the effect of lÂ°C warming may be advantageous in the spring and fall and disadvantageous in the summer. This tempers the net outcome of the changed weather. Finally, the b coefficients distilled from history indicate that the assumed climate change caused by C02 would reduce yields of the three crops in several states by 2 to l3%. However, a caution must be made concerning the calculated estimates of yield changes given in Tables 6.2, 6.3, and 6.4. These values are estimated on the basis of a uniform weather change of lÂ°C warmer and l0% drier throughout the whole year. It is unlikely that the warmer and drier climate brought about by an increase in atmospheric CO2 concentrations will be so evenly modulated through all the seasons nor from place to place. Also, we must assume that the killing extremes of weather are related to means in the same way in the changed climate as they are now. Nevertheless, the small percentage decreases estimated are remarkably consistent for the three major crops throughout the American grain belt. 6.3.2 Simulation The physiology of the crop and the physics of evaporation provide an alternative to history as a foundation for predicting the consequence of changed temperature and moisture. A well-known simulator of crop growth was composed by Duncan et al. (l967) . Essentially, they assembled such things as the photosynthesis of maize leaves as a function of radiation and temperature, the archi- tecture of the canopy of leaves, and the calendar of crop development. Other simulators have incorporated more details of metabolism, including the division of plant stuff between foliage and grain (Maas and Arkin, l980) . Still others have incorporated the transpiration of water and hence the depletion of soil water and occasional drought (Richardson and Ritchie, l973). Simulators have been composed for wheat (Maas and Arkin, l980) and soybeans (Curry, l97l). Although simulators are not perfect, they are logical assemblies of current physiology and physics of crops, and we shall examine the consequences of climatic change with the wheat simulator. The wheat simulator begins a season with parameters specifying a variety, date of planting, the water in the soil, and the soil's ability to hold water. Day by day the simulator uses the temperature, rainfall, and solar radiation to calculate use of water, plant growth, and its allocation to leaves and grain. If it becomes cold, evapora- tion, growth, and progress through the life cycle slows. If it rains, soil moisture is replenished. And if it becomes hot, growth and development decrease, evaporation increases, and soil moisture may be depleted, inhibiting growth. At the end of the season, the sum of the growth of grain is the yield, which reflects the variety and, especially, the weather. To perform our simulations we must have some weather observations. He began with the actual observations of nine crop reporting districts in North Dakota, each district with l5 to 20 stations. We chose the
404 Actual Weather Changed Weather (+1Â°C, -10% precip.) LU o til rr LL LU U oc a 20- 6.5 10.5 14.5 18.5 YIELD (q/ha) FIGURE 6.2 North Dakota simulated spring wheat yield (l949-l980) years l949 to l980. Because only temperature and rainfall were observed at these stations, the solar radiation required by the simulator was estimated from the temperature and precipitation (Richardson, l98l). We also needed a planting date for the simulated wheat yield, and we estimated it by the method of Hodges and Artley (l98l) in soil with a capacity for l75 mm of water. For the first spring of our simulations we assumed that the soil was filled to capacity, and in later springs the content was calculated from the rain and evaporation of the pre- ceding months. We used the parameters for a common variety of spring wheat in North Dakota. In the simulator an improvement in technology would be reflected in a change in some of the parameters. Thus, a better variety would change the productivity for the North Dakota variety. For each year we simulated the yield for each crop reporting dis- trict and added these to make a yield for North Dakota. The simulated yields for the 32 years, l949-l980, form a frequency distribution of 4.5 to l6.5 quintals/ha for the present climate (Figure 6.2). The distribution is skewed with a median yield of 8.5 quintals/ha. This is less than the actual average yield of l2.0 quintals/ha for the years l978-l980 (see Table 6.2). To eliminate the underestimation of the simulated yield would require tuning the model to provide values close to existing yields. We can, however, use the untuned model to calculate the relative change if rising CO2 makes the climate warmer and drier.
405 The weather for each crop reporting district was changed for increased (X>2 in the atmosphere by adding lÂ°C to the daily temperatures and decreasing each rain or snow fall by l0%. The lÂ°C warming increased evaporation and also the solar radiation, which we calculated from the temperature. The distribution of yields in the changed climate has a median yield of 6.5 quintals/ha (Figure 6.2). Both the median and range are 2 quintals/ha less than for the unchanged weather. The simulated change of 2 quintals/ha is more than the l.36 quintals/ha estimated from the b coefficients of Table 6.2. In relative terms the 2 is, of course, a substantially greater fraction of the median of 8.5 than l.36 is of its mean of l2 quintals/ha. Having to choose between the two, one chooses the actuality incorporated in the b coefficients and the ll% decrease and is pleased that the simulation is in the same direction and range. In addition, we have gotten a fre- quency distribution of yields from the simulator. In some years, the changed climate provided larger yields than the median for the normal climate, but in general the warmer and drier climate from a changed CO2 would decrease yield. 6.3.3 Summary The changes in yield that would follow a change to the warmer, drier climate assumed to follow an increase in atmospheric C02 have now been forecast by two orderly, explicit, and objective means. The reader knows we have not even employed every coefficient in our tables nor simulated wheat yields in every important region. Neither have we simulated corn or soybean yields. We have not combined the beneficial direct effect of C02 on plant growth with the disadvantages of warmer and drier, considered possible geographic shifts of the grain belt nor incorporated the changes breeders will make in varieties to exploit a new climate. We have considered only the single change of warmer and drier. Despite the fog created by the foregoing list of qualifications and warnings, a clear conclusion shines through. The warmer and drier climate assumed to accompany the increased CO2 will decrease yields of the three great American food crops over the entire grain belt by 5 to l0%, tempering any direct advantage of CO2 enhancement of photosynthesis. 6.4 PATHOGENS AND INSECT PESTS Sometimes a change in the weather that has only a modest direct effect on a crop is amplified into a disaster by a third partyâa pest. A concrete example of a plant disease (Tatum, l97l) makes the point. The spring of l970 was wet in the Southeast. An early hurricane striking Biloxi, Mississippi, exemplified the weather. The summer was also wet and warm in the Corn Belt. Summer moisture in the Corn Belt generally benefits corn, and a bumper crop should have been produced. Thanks to a third party, a fungal pest, the parties of hybrid corn and
406 moist weather did not produce the expected crop. It was, in fact, an eighth less than the year before, and yield per acre was off by one sixth. That was 535 million bushels of corn for the United States. Before a change in the weather can evoke an epidemic, a susceptible crop and a virulent pest must be on stage. In the United States, 70 million to 80 million acres are planted to a single host, corn; about l acre of cropland in 5 is planted to the single crop of corn. The host was uniform. When hybrid corn was introduced in the l930s, seed was produced by detasseling or emasculating the rows that were to be the female parent and allowing pollen to fly only from the rows designated as male parent. In l95l, geneticists invented a labor- saving scheme to obviate detasseling in the hot sun. They found in Texas a cytoplasm, T-cytoplasm, that could sterilize the tassels of the female parent, and they also found a restorer gene that the male parent could transmit to the offspring, letting it produce fertile pollen and a crop in the next year. The method was so profitable that by l970 most of the 67 million acres were planted to seed that inherited its cytoplasm from a single male-sterile plant from Texas. Despite the expanse of host, however, it was healthy. In a sampling of years from l935 to l969 only half as many reports were published about corn disease as about diseases of a crop grown on a much smaller acreage (Ullstrup, l972). The year l970 changed things. A new pest came on stage: Race T of Helminthosporium maydis. The microscopic fungi of the genus Helminthosporium have long caused only minor diseases of corn. Because all the Helminthospor ium diseases were accused of decreasing corn yield only about 2%, Americans naturally paid little attention when Philippine scientists reported in l96l an especial susceptibility of T-cytoplasm corn to Helminthosporium maydis. During l969, however, greater-than-normal susceptibility to Helminthosporium was observed in seed and test fields, and over winter the association between T-cytoplasm corn and Southern corn leaf blight, the disease caused by Helminthosporium maydis, was confirmed (Tatum, l97l). The pathogenic fungus had done what many shifty pathogens had done before: it had produced a new race that was selected for the vast acreage of uniform host, and it burst forward in the moist weather of l970. An urban society studied agriculture, the Chicago Tribune alone publishing 37 articles in l970 (Ullstrup, l972). A mutated microscopic fungus had amplified a change to a warm, moist summer, and the little fungus began a series of calamities that brought wheat deals, expensive groceries, and beef boycotts to the early l970s. Other diseases, like the wheat rust mentioned in Section 6.l, also amplify the effect of changed weather. The pests of insects and weeds are other amplifiers. Whereas the encouragement of grasshoppers by drought cited in Section 6.l is straightforward, some amplifications are more convoluted. Aphids carry viruses that infect crops, and because aphids usually multiply faster and move more frequently in warm weather than in cool, viruses spread faster in southern potato fields, and seed certified to be virus-free is grown in northern Maine.
407 Biological control provides another convoluted case. The walnut aphid attacked orchards in both the warm, dry San Joaquin Valley and cooler coastal valleys. A parasitic insect from France controlled the walnut aphid in the cool coastal valleys, but an Iranian biotype of the parasite was required to control the aphid pest on walnuts in the warmer Valley (Eraser and van den Bosch, l973). Being photosynthesizing plants, weeds can be directly affected by CO2 as described above. They can also be affected by changes in the weather and amplify the effect on crops. A "pestograph" (U.S. Dept. of Agriculture, l970; National Research Council, l976) shows how. Field bindweed and Canadian thistle compete well with crops in cool, dry climates. If the climate is moister, the thistle persists and is joined by quackgrass. If instead the climate is both moister and warmer, both thistle and bindweed suffer and foxtail, coffeeweed, and nutsedge prosper. Adding a convolution, we mention that the pesticides for controlling diseases, insects, and weeds are affected by the weather, especially moisture. Moisture may affect the absorption of a pesticide, particu- larly an herbicide, or rain may simply wash it away. Because pest outbreaks require the threesome of virulent pest, sus- ceptible crop, and favorable weather, outbreaks cannot be predicted from weather alone. Many seasons of favorable weather passed before the Southern corn leaf blight epidemic of l970 or the tobacco blue mold epidemic of l979 (Lucas, l980) because another factor was missing. Nevertheless, week-to-week changes in pests can be anticipated from the weather, and mathematical models of the effect of weather on pests have been composed. These range from statistical regressions between weather and disease on wheat (i.e., Burleigh et al., l972) to simula- tions on a computer of how weather affects the stages in the life of a pathogen (Waggoner et al., l972). Some simulations even unite a simulator of a weevil with one of alfalfa (Miles et al., l974). The practical goal of such models is generally more precise timing of pesticides, no needless application, and fewer pests. Given a pest to consider, however, the models can show how a pest will change with the weatherâif new actors do not enter. History shows, however, that pests are so varied and variable that new and surprising pests will appear so frequently that a quantitative calculation for a given pest would give a misleading certainty. Instead, agriculture must expect novel pests, keep its research system in trim, and control new pests as they arise or the direct and indirect effects discussed above will be rendered trivial by a shifty and aggressive pest. 6.5 IRRIGATION IN A WARMER AND DRIER CLIMATE Irrigation is both important and susceptible. Its importance derives from its expanse and from the value of irrigated crops. Fully 50 million acres or about l in 7 American acres of cropland are irrigated. The quarter trillion cubic meters of irrigation water withdrawn from American streams and groundwater represents about half of all with- drawals of this natural resource. Averaging wheat yields over all
408 American fields, humid as well as arid, one sees an average of l.9 ton/ha on unirrigated versus 3.7 ton/ha on irrigated land. Valuable crops are grown on irrigated fields because irrigation reduces vari- ability of water and produces consistently high yields. Thus, most of the irrigated cropland (44 million acres) occurs on only l2% of the farms, and these farms produce fully 40% of the market value of the crops from all American cropland! (Jensen, l982b.) Since irrigation uses runoff and runoff is a fraction of precipita- tion, the effect of a prolonged change in the amount of precipitation will usually be a greater change in the runoff to supply irrigation. For example, a simulation of a l0% decrease in precipitation over a 2000-km2 watershed with an annual mean runoff of 400 mm showed a decrease in runoff of 25% to 300 mm per year. Furthermore, a l0% decrease in precipitation on a 9000-km2 watershed with only ll-mm mean runoff would decrease runoff 40% to 7 mm per year (J. Nemec and J. Schaake, l982, "Sensitivity of Water-Resource Systems to Climate Variation," unpublished manuscript, WMO Secretariat, Geneva). This reasoning cannot, however, be applied blindly to a large watershed. For example, the runoff in the bulk of the Colorado River Basin is little, but in the high Rocky Mountains, which contribute most of the water, runoff is a larger portion of the precipitation, and a l0% reduction in precipitation could decrease runoff more nearly l0% than the preceding calculations suggest. This is treated in Chapter 7. The agricultural effects from changes in precipitation patterns reducing runoff for irrigation might be mitigated by transferring water from one river basin to another. Southern California's agriculture is largely sustained by such large diversions of water. But such large water storage and transfer projects take many decades to accomplish and depend on an accurate prediction of the changes in the total runoff from the effected river basins (Cooper, l982). The decline of irrigation systems, often caused by salinity and rising water table, is not a new experience for mankind and intrigues archaeologists studying the extinction of cultures. Even without drastic changes in the availability of water, irrigated land faces a limited period of production because of the buildup of residual salts carried into the land by the irrigation water and left behind as water evaporates or is transpired into the atmosphere. Of the l450 x l0 ha of world cropland under agricultural cultivation about l6% (230 x l06 ha) is irrigated and about 2 x l05 ha of this irrigated land is removed from production annually because of buildup of salt concentra- tions that inhibit crop production (Hodges et al., l98l). Without increased availability of water to wash away salts, irrigated land is ultimately lost to crop production. Salt-tolerant species can be developed as new crops or standard crops can be bred to be salt tolerant; but finally all crops fail in excessively salty soil. Right now in America, we foresee another sort of decline without a change in the weather: Extraction of groundwater exceeds recharge by 25 billion to 30 billion cubic meters annually. About 60% of this overdraft is in the Ogallala aquifer, and by A.D. 2000 the conversion from irrigated to dryland farming is expected with the accompanying years of fallow and changes in crops and yields (Jensen, l982b).
409 The loss in irrigated yield accompanying the change in weather specified earlier can be envisaged in two ways. For grain one might simply and roughly say that some areas can no longer be irrigated, and on those acres the yield will be halved at least because the subsequent dryland crops will be grown in alternate years. The value of truck crops would sustain their irrigation. In l977, California and Texas produced about half the value of the principal commercial vegetable crops. These two states have a third of the irrigated cropland, and a decrease in water and irrigated area in those states could reduce yields to zero on many acres and thus decrease the fresh vegetables in the produce section of the supermarket, especially in the winter. 6.6 ADAPTING TO THE CHANGE TO A WARMER, DRIER CLIMATE The predictions in the preceding sections tell from a foundation of physiology and history how a change in climate would change the yields of crops if farmers persisted in planting the same varieties of the same species in the same way in the same place, ignoring the weather. The histories related in Section 6.l, however, are filled with stories of adaptations; and with more mobility, technology and knowledge, the future farmers will surely adapt even faster than the ones of the histories. The safest prediction of any we shall make is: Farmers will adapt to a change in climate, exploiting it and making our preceding predictions too pessimistic. If the climate changes, farmers will move themselves, change crops, modify varieties, and alter husbandry. The loss of cropland to the margins of the desert, for example, may be replaced in the national production by higher yields and more cropland at the frigid margins. Seeking higher yields and more profit, they will correct their course annually, and they may even adapt to a slowly changing climate unconsciously and successfully. The seasonal adaptation by migration between the marginal climates near deserts or mountains is familiar, although the movable animals are not the stationary crops that have been the chief concern in this chapter. In Section 6.l permanent migrations were exemplified by the flight from the Middle Border in the l890s and from the Dust Bowl in the l930s. With the facility of modern communication and the scope of a nation that would still span many climates, farmers will surely move as promptly as the Okies, saving themselves while abandoning the cropland to some other use. 6.6.1 Breeding New Varieties The change of crop species was exemplified in Section 6.l by the change from corn to wheat as Kansas grew drier from l885 to l905. Although the market rather than the weather was the impetus, the rapidity of the adoption of new species of crops was recently demonstrated in Minnesota and North Dakota (Frey, l982): from l976 to l980 the acreage of sun-
4l0 flowers in the two states increased nearly sixfold to 2 million ha, and then in 2 more years it decreased by a third as dry beans and corn became more profitable than sunflowers in the Red River Valley. In the process, the area of small grains decreased and has not recovered. Adaptation by a change of crop can be swift, and a change in CO2 and climate will make an opportunity for introducing new species in farming and make the preservation of species and the introduction of plants more important. Changing the variety of a crop by planting a different seed can be even swifter than changing crops because so little needs to be altered from the dealer who supplies chemicals to the farmer who must finance equipment on to the consumer who may scarcely realize the change. The question is whether breeders can make new varieties adapted to a climate as fast as it changes. The adaptability of crops was long ago demonstrated. In l857, Wendeling Grimm brought his family and 7 kg of alfalfa seed from The Grand Duchy of Baden to Minnesota. Although the Minnesota winters, more severe than the German winters, killed most of Grimm's alfalfa, some survived. Planting seed from the survivors again and again, he gradually developed an alfalfa that could become relatively dormant in the fall and resist Minnesota cold (Burton, l980). Given time for population improvement, Grimm produced a new variety for a sudden change to much colder winters. Winter wheat is another example (Rosenberg, l982). Since the l920s the growing zone for hard red winter wheat has been expanded through breeding and improved agronomic technologies to include climates with a range of temperatures and rainfall comparable with or greater than those predicted to occur with a doubling of the atmospheric CO2 concentration. Because new varieties of crops are bred and tested outdoors, they are indirectly adapted. For example, the yield of modern hybrids of corn in seven states during l972-l976 decreased only a third in bad years, whereas the yield of open-pollinated corn in the seven states during l928-l936 fell fully two thirds in bad years (Harvey, l977). When one open-pollinated cultivar and 24 corn hybrids developed during l930-l970 were compared, the new hybrids were especially superior in environments that caused low yields because they were less sensitive to drought (Russell, l974). The breeding of adapted varieties does have limits and costs. The forage called Bermuda grass tolerates heat and drought, and its toler- ance of cold has been improved. Still, it is restricted to central and southern states (Burton, l980). Tolerance for one stress may bring susceptibility to another as when plants selected for heat and drought tolerance do not have early vigor in cold soil (Jensen, l98l). Still, some costs can become benefits: Since cold tolerance is correlated with low yields (Jensen, l98l), it must also be said that adapting varieties to a warmer climate will bring higher yields. Plant breeding must keep pace with the change in climate if yields and the food supply are not to decline. Although knowing the factor that has changed in the environment, learning the mechanism of resis- tance to the changed factor, and then rationally engineering just that change in the genes of the crop is appealing, the process is still beyond our capabilities. We are not, however, vulnerable.
4ll Since plant breeding and testing is performed in the open and in the region of the intended farming, the logical but lengthy process of measuring the weather, discovering the mechanism of resistance, and engineering the genes is truncated and accelerated. For example, a single producer of seed corn has 22 testing sites in the United States and l7 abroad where a quarter million rows of breeding material and 400 inbred lines are yield tested. If the climate shifts, some of this multitude will perform well in the new environment and be shared among sites and advanced without time lost to follow the logical but lengthy chain. Although a complete breeding cycle is approximately a decade from the beginning of inbreeding to the marketing of a product, the objectives can be shifted during the first 5 years, and new hybrids emerging are those adapted to the final 3 years (Frey, l982). For maize and other major crops, other seed producers and experiment stations provide the insurance of other programs and diversity. Although fads and excursions present some hazards, steady work by the plant-breeding establishment will surely produce varieties of the life-sustaining, major crops that sustain man that are well adapted to the environment as it shifts gradually to lÂ°C warmer and l0% drier than the present. 6.6.2 Adapting to Less Water The narrowing of stomata by rising C02 may reduce the transpiration of water. Reducing the harm of drought is, however, a well-known game played by storing more precipitation and getting more yield from the stored water. Much that follows is taken from Greb (l979). Increasing storage is exemplified by a history of cropping in the Great Plains (see Table 6.5). Crops are planted every other year to permit soil moisture to build up during the fallow year to support such alternate year cropping. Tillage during the fallow year controls weeds and decreases the loss of soil moisture while stubble captures more TABLE 6.5 Progress in Fallow Systems and Wheat Yields, Akron, Colorado (Greb, l979) Fallow Efficiency Storage/ Water-Use Precipitation Yield Efficiency Years Fallow System (%) (tons/ha) (kg/m3) l9l6-l930 Plow harrow l93l-1945 Shallow disk, rod weeder l946-l960 Improved conventional tillage, begin stubble mulch l957 l96l-l975 Stubble mulch, begin minimum tillage with herbicides l969 l976-l990 Estimate. Begin no-till l983 l9 24 27 33 40 l.l l.2 l.7 2.2 2.7 0.l2 0.l4 0.2l 0.28 0.32
412 precipitation. Together these doubled the proportion of precipitation stored, which is called the fallow efficiency. Other means of increas- ing storage include barriers that catch snow, leveling and terracing to decrease runoff, and harvesting water from nearby acreage. Controlling nonbeneficial plants along western rivers could increase water supply by 3l billion nr or about the same as the overdraft of groundwater in America (Jensen, l982a). Matching irrigation to need can decrease pumping, but if the former excess water was flowing back to the water supply, the saving of water will be moderated. Getting more yield per acre increases the yield of marketable product per unit of water consumed, which is called water-use efficiency. Gen- erally, evaporation from a productive crop is little more than from a crop that merely shades the ground. The increase in water-use effi- ciency can be seen in Table 6.5 where yields as well as storage increase with time, increasing water-use efficiency even more than fallow efficiency. Changing the crop as well as increasing fertility can increase water-use efficiency, Table 6.6. The use of water in this example is changed insignificantly by either plant or fertilizer. Since the yield of grass is tripled from the least fertilizer on the less productive grass to the most fertilizer on the more productive grass, the water-use efficiency is also tripled. In the same way, breeding more productive varieties and controlling pests increase water-use efficiency. A final and large adaptation is moving to a different place. From l944 to l978 the irrigated cropland in California increased l480 thousand ha. At about the same time the California cropland of commercial vegetables for fresh market increased 84 thousand ha, while that in the Atlantic states from North Carolina to Georgia decreased 20 thousand ha. During the same period, the area in New York and New Jersey declined by 57 thousand ha. This was an adaptation by moving to the warmth, sun, and irrigation of California. If water becomes short in California, the nation might adapt by increasing its vegetable fields near the Atlantic where the rain falls. TABLE 6.6 The Effect of Plant and Fertilizer on Water-Use Efficiency, Akron, Colorado (Greb, l979) Water-Use Nitrogen Water Use Yield Efficiency Plant (kg/ha) (mm) (tons/ha) (kg/m3) Russian wild rye Crested wheatgrass 0 28 56 0 28 56 l59 l94 l6l l58 l59 l56 0.95 .54 .84 .76 2.22 2.72 l. l. l, 0.56 0.75 l.08 l.06 l.33 l.64
4l3 6.7 CONCLUSION We have called the roll of direct effect of CO2 on plants and of side effects of warmer and drier, pests, and adaptation. What will be the net or integration of all of these on yield? Although answering seems foolhardy rather than courageous, some large matters seem clear. The direct effects of more CO2 in the air are beneficial: increase CO2 around a prosperous leaf and it will assimi- late more carbon and lose less water. The indirect effects of warmer and drier, on the other hand, are slightly harmful in the American grain belt as calculations from both past statistics and physiological simu- lators show. Although pests will change, the direction that they will change is imponderable. While CO2 is directly narrowing stomata and the need for water, it is also decreasing rainfall. This conservative forecast of countervailing effects influences the prediction of CO2 in the air. A portion of each ton of CO2 injected into the atmosphere from a stack or exhaust can shortly dissolve in the ocean or become sugar and eventually wood or soil organic matter, tempering the increase in atmospheric CO2. If real plants in corn fields and forests increase their net photosynthesis along the curves observed in the laboratory, the increase in atmospheric CO2 would be tempered by plants to about 500 ppmv. Unfortunately, to the extent that indirect effects of warmer and drier prevent the rise in yield expected from the laboratory experiment with CO2, vegetation will temper the rise in CO2 to a lesser extent. Thus in the end one sees that the effects on plants of the gradual changes in CO2 and gradual changes in weather foreseen for A.D. 2000 are modest, some positive and some negative. The wise forecast of yield, therefore, seems a continuation of the incremental increases in production accomplished in the past generation as scientists and farmers adapt crops and husbandry to an environment that is slowly changing with the usual annual fluctuations around the trend. REFERENCES Akita, S., and D. N. Moss (l973). Photosynthetic responses to CO2 and light by maize and wheat leaves adjusted for constant stomatal apertures. Crop Sci. l3:234-237. Baker, D. N. (l962). The effect of air temperature on the rate of photosynthesis in corn. Ph.D. Thesis, Cornell University, Ithaca, N.Y. Baker, D. N. (l965). Effects of certain environmental factors on net assimilation in cotton. Crop Sci. 5:53-56. Barnhart, J. D. (l925). Rainfall and the Populist party in Nebraska. Am. Political Sci. Rev. l9:527-540. Bjorkstrom, A. (l979). Model of CO2 interactions between atmosphere, oceans and land biota. In The Global Carbon Cycle, B. Bolin et al., eds. Wiley, New York, pp. 403-457. Brooks, C. E. P. (l970). Climate Through the Ages. 2d ed. Dover, New York.
4l4 Burleigh, J. R., M. G. Evetsmeyer, and A. P. Roelfs (l972). Develop- ment of linear equations for predicting wheat leaf rust. Phyto- pathology 62:947-953. Burton, G. W. (l980). Breeding programs for stress tolerance in forage and pasture crops. Paper presented at Workshop on Crop Reactions to Water and Temperature Stress Tolerance in Forage and Pasture Crops, Duke U., Durham, N.C. Chandler, J. M. (l98l). Estimating losses of crops to weeds. Handbook of Pest Management in Agriculture, Vol. l, D. Pimentel, ed. CRC Press, Boca Raton, Fla., pp. 95-l09. Cooper, C. F. (l982). Food and fiber. In Carbon Dioxide Review: l982, W. C. Clark, ed. Oxford U. Press, New York, pp. 297-320. Curry, R. B. (l97l). Dynamic simulation of plant growth, Part I. Development of a model. Trans. Am. Soc. Agric. Eng. l4:946-949. Duncan, W. G., R. S. Loomis, W. A. Williams, and R. Hanau (l967). A model for simulating photosynthesis in plant communities. Hilgardia 18:l8l-205. Enoch, H. Z., I. Rylski, and Y. Samish (l970). CO2 enrichment to cucumber, lettuce and sweet pepper plants grown in low plastic tunnels in a subtropical climate. Israel J. Agric. Res. 20:63-69. Enoch, H. Z., I. Rylski, and M. Spigelman (l976). C02 enrichment of strawberry and cucumber plants grown in unheated greenhouses in Israel. Sci. Hortic. 5:33-4l. Evans, L. T. (l975). The physiological basis of crop yield. Crop Physiology, L. T. Evans, ed. Cambridge U. Press, Cambridge, pp. 327-335. Fraser, B. D., and R. van den Bosch (l973). Biological control of the walnut aphid in California: the interrelationship of the aphid and its parasite. Environ. Entomol. 2:56l-548. Frey, N. M. (l982). Plant breeding under increasing C02 concentra- tion and in a changing climate. Paper presented at AAAS Conference on Plants under Increasing C02. Athens, Ga. Gifford, R. M. (l979). Growth and yield of CO2-enriched wheat under water-limited conditions. Aust. J. Plant Physiol. 6:367-378. Gifford, R. M., P. M. Bremner, and D. B. Jones (l973). Assessing photosynthetic limitation to grain yield in a field crop. Aust. J. Agric. Res. 24:297-307. Greb, B. W. (l979). Reducing drought effects on croplands in the West Central Great Plains. USDA Agr. Info. Bull. 420. Guinn, G., and J. Mauney (l980). Analysis of CO2 exchange assumptions: feedback control. Predicting Photosynthesis for Ecosystem Models, Vol. 2, J. D. Hesketh and J. W. Jones, eds. CRC Press, Boca Raton, Fla., pp. l-l6. Haigh, P. A. (l977). Separating the Effects of Weather and Management on Crop Production. C. F. Kettering Foundation (ST 77-4), Dayton, Ohio. Hardman, L. Â£., and W. A. Brun (l97l). Effect of atmospheric CO2 enrichment at different developmental stages on growth and yield components of soybeans. Crop. Sci. ll:886-888. Hardy, R. W. F., and U. D. Havelka (l977). Possible routes to increase the conversion of solar energy to food and feed by grain legumes and
4l5 cereal grains (crop production): C02 and N2 fixation, foliar fertilization, and assimilate partitioning. In Biological Solar Energy Conversion, A. Mitsui et al., eds. Academic, New York, pp. 299-322. Harlan, J. R. (l975). Crops and Man. American Society of Agronomy, Madison, Wise. Harvey, P. H. (l977). ARS National Research Program, NEP No. 20040 breeding and production: corn, sorghum and grain millets. Am. Seed Trade Assoc. l977 Yearbook and Proceedings 94th Annual Conference, pp. l78-l82. Hesketh, J. D. (l963). Limitations to photosynthesis responsible for differences among species. Crop Sci. 3:493-496. Hodges, C. N., R. M. Fontes, E. P. Glenn, S. Katzen, and L. B. Colvin (l98l). Sea water-based agriculture as a food production defense against climate-variability. In Food-Climate Interactions, W. Bach et al., eds. Reidel, Dordrecht, Holland, pp. 8l-l00. Hodges, R., and J. A. Artley (l98l). Spring Small Grains Planting Date Distribution Model. AgRISTARS Rep. SR-Ll-040332. March l98l, JSC l6858. Lyndon B. Johnson Space Center, Houston, Tex. Jensen, M. E. (l982a). Water resource technology and management. Proceeding RCA Symposium: Future Agricultural Technology and Resource Conservation. December 5-9, l982, Washington, D.C. Jensen, M. E. (l982b). Overview-irrigation in U.S. arid and semiarid lands. Prepared for the Office of Technology Assessment, Water-Related Technologies for Sustaining Agriculture in Arid and Semiarid Lands. Jensen, R. G., and J. T. Bahr (l977). Rubulose l,5-bisphosphate carboxylase-oxygenase. Ann. Rev. Plant Physiol. 28:379-400. Jensen, S. D. (l98l). Discussion. In Plant Breeding II, K. J. Frey, ed. Iowa State U. Press, Ames, p. l7l. Kellogg, W. W. (l977). Effects of Human Activities on Global Climate. Technical Note No. l56. WHO-No. 486, World Meteorological Organization, Geneva. Kimball, B. A. (l982). Carbon dioxide and agricultural yield: an assemblage and analysis of 430 prior observations. WCL Report ll. U.S. Water Conservation Laboratory, Phoenix, Ariz., pp. l-48. Lawyer, A. L., K. L. Cornwell, P. O. Larsen, and J. A. Bassham (l98l). Effects of carbon dioxide and oxygen on the regulation of photo- synthetic carbon metabolism by ammonia in spinach mesophyll cells. Plant Physiol. 68:l23l-l236. LeDuc, S. K. (l980). Corn models for Iowa, Illinois and Indiana. NOAA Center for Environmental Assessment Services, Columbia, Mo. Lemon, E. R., ed. (l983). CO? and Plants: The Response of Plants to Rising Levels of Atmospheric Carbon Dioxide. AAAS Selected Symposium No. 84. Westview Press, Boulder, Colo. Lucas, G. D. (l980). The war against blue mold. Science 2l0:l47-l53. Maas, S. J., and G. F. Arkin (l980). TAMW: A wheat growth and development simulation model. Program in Model Documentation Report No. 80-3, October l980. Blackland Research Center, Texas Agricultural Experiment Station, Temple, Tex.
4l6 Malin, J. C. (l936). The adaptation of the agricultural system to sub-humid environment. Agric. History l0:ll8-l4l. Manabe, S., and R. J. Stouffer (l980). Sensitivity of a global climate model to an increase of CO2 concentration in the atmosphere. J.. Geophys. Res. 85:5529. Manabe, S., R. T. Wetherald, and R. J. Stouffer (l98l). Summer dryness due to an increase of atmospheric C02 concentration. Clim. Change 3_:347-386. Mauney, J. R., K. E. Fry, and G. Guinn (l978). Relationship of photo- synthetic rate to growth and fruiting of cotton, soybean, sorghum, and sunflower. Crop Sci. l8:259-263. McCullough, D. (l98l) Mornings on Horseback. Simon & Schuster, New York, 445 pp. McQuigg, J. D. (l979). Climatic variability and agriculture in the temperate regions. In WHO Proc. World Climate Conference, WMO-No. 537, World Meteorological Organization, Geneva, pp. 406-425. Miles, G. E., R. M. Peart, R. J. Bula, M. C. Wilson, and T. R. Hintz (l974). Simulation of plant-pest relationships with GASP IV. ASAE Paper 74-4022. American Society of Agricultural Engineers, St. Joseph, Mich. Mitchell, J. M., Jr. (l977). Carbon dioxide and future climates. EDS March. Environmental Data Service, National Oceanographic and Atmospheric Administration, Washington, D.C., pp. 3-9. Motha, R. P. (l980). Soybean models for Iowa, Illinois and Indiana. NOAA Center for Environmental Assessment Services, Columbia, Mo. National Research Council (l976). Climate and Food: Climatic Fluctuation and U.S. Agricultural Production. Board on Agriculture and Renewable Resources, Commission on Natural Resources, National Academy of Sciences, Washington, D.C. Neild, R. E., H. N. Richman, and M. W. Seeley (l979). Impacts of different types of temperature change on the growing season for maize. Agric. Meteorol. 20:367-374. Patterson, D. T. (l982). Effects of light and temperature on weed/crop growth and competition. In Biometeorology in Integrated Pair Management, J. L. Hatfield and I. J. Thomason, eds. Academic, New York, pp. 407-420. Patterson, D. T., and E. P. Flint (l980). Potential effects of global atmospheric C02 enrichment on the growth and competitiveness of Cj and C4 weed and crop plants. Weed Sci. 28:7l-75. Patterson, D. T., and E. P. Flint (l982). Interacting effects of C02 and nutrient concentration. Weed Sci. 30:389-394. Richardson, C. W. (l98l). Stochastic simulations of daily precipita- tion, temperature, and solar radiation. Water Resources Res. r7:l82-l90. Richardson, C. W., and J. T. Ritchie (l973). Soil-water balance for small watersheds. Trans. Am. Soc. Agric. Eng. l6:72-77. Rogers, H. H., et al. (l980). Response of Vegetation to Carbon Dioxide. No. 00l. Field Studies of Plant Responses to Elevated Carbon Dioxide Levels. Research Progress Report from the U.S. Department of Energy and the U.S. Department of Agriculture, DOE Carbon Dioxide Research Division, Washington, D.C. 20545.
4l7 Rosenberg, N. J. (l982). The increasing C02 concentrations in the atmosphere and its implications on agricultural productivity, II. Effect through CO2-induced climate change. Clim. Change 4:239-254. Russel, W. A. (l974). Comparative performance for maize hybrids representing different eras of maize breeding. Proceedings 29th Annual Corn and Sorghum Res. Conf., Am. Seed Trade Association, pp. 8l-l0l. Sakamoto, C. (l978). Reanalysis of CCEAl, U.S. Great Plains Wheat Yield Models. Center for Climatic and Environmental Assessment Technical Note No. 78-3. National Aeronautics and Space Adminis- tration, Houston, Tex. Sakamoto, C. (l982). Application of Process Models for Large Area Crop Yield Estimation. Level I AgRISTARS Project Review, April l9, l982. Lyndon B. Johnson Space Center, Houston, Tex. Schlebecker, J. T. (l953). Grasshoppers in American agricultural history. Agric. History 27:85-93. Sionit, N., H. Helmers, and B. R. Strain (l980). Growth and yield of wheat under CO2 enrichment and water stress. Crop Sci. 20:687-690. Sionit, N., D. A. Mortensen, B. R. Strain, and H. Helmers (l98la). Growth response of wheat to C02 enrichment and different levels of mineral nutrition. Agron. J. 73:l023-l027. Sionit, N., B. R. Strain, and H. A. Beckford (l98lb). Environmental controls on the growth and yield of okra, I. Effects of temperature and of CO2 enrichment at cool temperature. Crop Sci. 2l:885-888. Steinbeck, J. (l939). Grapes of Wrath. Viking, New York. Tatum, L. A. (l97l). The Southern corn leaf blight epidemic. Science JL7l:lll3-lll6. Thompson, L. M. (l969) Weather and technology in the production of corn in the U.S. Corn Belt. Agron. J. 6l:453-456. Tuchman, B. (l978). A Distant Mirror; The Calamitous l4th Century. Knop, New York. Ullstrup, A. J. (l972). Impacts of Southern corn leaf blight epidemics of l970-l97l. Ann. Rev. Phytopathol. l0:37-50. U.S. Dept. of Agriculture (USDA) (l970). Selected weeds in the United States. Agricultural Handbook 366. U.S. Government Printing Office, Washington, D.C. U.S. Dept. of Agriculture (USDA) (l972). Extent and Cost of Weed Control with Herbicides and an Evaluation of Important Weeds, l968, USDA ARS-H-l. U.S. Department of Agriculture, Washington, D.C. U.S. Dept. of Agriculture (USDA) (l982). Agricultural Statistics, l98l. U.S. Government Printing Office, Washington, D.C. van Veen, J. A., and E. A. Paul (l98l). Organic carbon dynamics in grassland soils. I. Background information and computer simulation. Can. J. Soil Sci. 6l:85. Waggoner, P. E., J. L. Monteith, and G. Szeicz (l964). Decreasing transpiration of field plants by chemical closure of stomata. Nature .20l:97-98. Waggoner, P. E., J. G. Horsfall, and R. J. Lukens (l972). EPIMAY: A simulator of Southern corn leaf blight. Conn. Agric. Exp. Stn. Bull. 729. New Haven, Conn.
4l8 Walker, D. A. (l976). Plastids and intracellular transport. In Transport in Plants III, C. R. Stocking and U. Heber, eds. Encyclopedia of Plant Physiology. Springer-Verlag, Berlin, pp. 85-l36. Wong, S. C. (l979). Elevated atmospheric partial pressure of CO2 and plant growth. Oecologia 44:68-74. Zelitch, I. (l97l). Photosynthesis, Photorespiration, and Plant Productivity. Academic, New York. Zelitch, I. (l982). The close relationship between net photosynthesis and crop yield and the critical role of oxygen stress. BioScience 12:796-802. Zieslin, N., A. H. Halavy, and H. Z. Enoch (l972). The role of CO2 increasing the yield of "Baccara" roses. Hortic. Res. l2:92-l00.