What Decreases Carbon Dioxide Levels in the Atmosphere

Suggested Citation:"7. Effects of Increasing Carbon Dioxide Levels and Climate Change . . .." National Research Council. 1991. Managing Water Resources in the West Under Conditions of Climate Uncertainty: A Proceedings. Washington, DC: The National Academies Press. doi: 10.17226/1911.

7
Effects of Increasing Carbon Dioxide Levels and Climate Change on Plant Growth, Evapotranspiration, and Water Resources

Leon Hartwell Allen, Jr.

U.S. Department of Agriculture

Gainesville, Florida

The atmospheric carbon dioxide concentration has risen from about 270 parts per million (ppm) before 1700 to about 355 ppm today. Climate changes, including a mean global surface temperature rise of between 2.8 and 5.2°C, have been predicted by five independent general circulation models (GCMs) for a doubling of the carbon dioxide concentration. The objectives of this paper are to examine plant responses to rising carbon dioxide levels and climatic changes and to interpret the consequences of these changes on crop water use and water resources for the United States.

BACKGROUND: PLANT RESPONSES TO ENVIRONMENTAL FACTORS

The main purpose of irrigation is to supply plants with adequate water for transpiration and for incorporating the element hydrogen in plant tissues through photosynthesis and subsequent biosynthesis of various tissues and organs. Transpirational flux requires several hundred times more water than photosynthesis.

In a series of U.S. Department of Agriculture studies beginning in 1910 in Akron, Colorado, Briggs and Shantz (1913a,b; 1914) showed that the water requirement of plants is linearly related to the biomass production of plants. They established this linear relationship by growing plants in metal containers filled with soil. Throughout the period of growth, they monitored water use carefully by weighing and adding measured amounts of water to maintain a desirable soil water content as water lost by plant transpiration was replenished.

The findings of Briggs and Shantz have been confirmed repeatedly (Allison et al., 1958; Arkley, 1963; Chang, 1968; Hanks et al., 1969; Stanhill, 1960). Figure 7.1 shows the linear relationship between biomass produced and rainfall plus irrigation water used by Sart sorghum and Starr millet in Alabama, as adapted from data of Bennett et al. (1964). De Wit (1958) examined the relationships among climatic factors, yield, and water use by crops. He found the following general linear relationship to be true, especially in semiarid climates:

where

Y = yield component (e.g., total above-ground biomass or seed production)

T = cumulative actual transpiration

T_max = maximum possible cumulative transpiration

m = constant dependent on yield component and species, especially on differences among photosynthetic mechanisms

Pan evaporation was used to represent T_max, which is proportional to climatic factors, especially air vapor pressure deficit (VPD):

where

e_s = the saturation vapor pressure at a given air temperature

e_a = the actual vapor pressure that exists in the air.

Combining these relationships, we see that yield is proportional to cumulative transpiration divided by vapor pressure deficit:

where k is a constant with units millibars • g (dry matter) • g^-1 (water). Like m, k depends on yield component, species, and photosynthetic mechanisms.

FIGURE 7.1 Linear relationship between biomass production and water use for two forage crops in 1956 and 1957 at Thorsby, Alabama. Squares: Sart sorghum. Triangles: Starr millet.

SOURCE: Adapted from Bennett et al., 1964.

Thus, we can see that theory predicts that yield will be proportional to cumulative transpirational water use, divided by vapor pressure deficit. There are several ways of calculating the VPD; it can be computed by aggregating seasonal daytime average VPD, or by using approximation methods based on daily maximum and minimum temperatures (Jensen, 1974). As pointed out by Tanner and Sinclair (1983), the maximum es can be computed from the daily maximum temperature, and e_a can be estimated from the daily minimum temperature. Tanner and Sinclair estimated that the effective daytime e_s falls at a point two-thirds to three-quarters of the distance between the e_s computed at the daily maximum temperature and the e_a computed at the daily minimum temperature. The effective daytime VPD values then must be averaged over the growing season of the crop. Regardless of the method used to compute a representative VPD, yield versus cumulative transpiration linear relationships vary with the aridity of the

climate—specifically with the temperature and vapor pressure regime under which the crop is grown. Figure 7.2 (modified from Stanhill, 1960) shows water used versus dry matter yield of pastures from the latitude of Denmark (which has a cool, humid atmosphere) to the latitude of Trinidad (which has a hot, dry atmosphere). Based on comparisons among existing climates, we can expect that transpirational water requirements of plants will increase if climates get warmer.

Atmospheric carbon dioxide is known to affect plant yield. Kimball (1983) reviewed 430 observations of carbon dioxide enrichment studies conducted prior to 1982 and reported an average yield increase of 33 percent, plus or minus 6 percent, for a doubling of the carbon dioxide concentration. This value has been generally confirmed by many other studies since that time. The yield increases seem to apply for both biomass accumulation and grain yield. Thus, plants may grow larger and, considering Figure 7.1, they may use more water as the global carbon dioxide concentration increases.

Transpirational water use is clearly related to ground cover (Jensen, 1974; Doorenbos and Pruitt, 1977). Daily water use soon after crops are planted on bare soil is typically only 10 to 20 percent of water use after effective ground cover is reached. Water use rises sharply as the crop's leaf area increases. Similarly, water use drops 60 to 70 percent when hay crops such as alfalfa are cut. As leaf regrowth occurs, transpiration rates recover rapidly as the ground cover of leaves is restored. Ground cover can be quantified with a leaf area index (LAI): the ratio of leaf area per unit ground area. Therefore, any carbon dioxide-induced stimulation of early growth of leaf area or increase of total leaf area growth may increase transpiration.

Increased carbon dioxide concentrations are known to cause smaller stomatal apertures and hence to decrease the leaf conductance for water vapor (Morison, 1987). This is a second mechanism whereby increased carbon dioxide concentrations may affect plant transpiration.

Another effect of rising carbon dioxide concentrations is the change in water-use efficiency (WUE). Water-use efficiency has a range of definitions. For whole-season processes, it is best defined as the ratio of dry matter (or seed yield) produced to the amount of water used by crops. For shorter-term whole canopy processes, it is best defined as the ratio of the photosynthetic carbon dioxide uptake rate per unit land area to the transpiration rate per unit land area. Figure 7.2 demonstrates the effect of climate on WUE.

FIGURE 7.2 Cumulative dry matter yield versus cumulative potential evapotranspiration (ET) of pastures under a range of climatic regimes. Open circle: Denmark. Filled circle: The Netherlands. Open triangle: England. Filled triangle: New Jersey. Open Square: Toronto, Canada. Filled square: Gilat, Israel. Open inverted triangle: Trinidad, West Indies.

SOURCE: Adapted from Stanhill, 1960.

Equation 3 quantifies the relationship between WUE and vapor pressure deficit.

In summary, the following relationships have been established by research:

Transpiration is linearly related to biomass accumulation and yield.
Transpiration is also linearly related to the aridity of the climate—in other words, to the vapor pressure deficit. Thus, rising global temperatures would increase transpiration by increasing the atmospheric vapor pressure deficit.

Transpiration is affected by the degree of ground cover.
Rising carbon dioxide concentrations will increase plant growth. More rapid leaf area development and more total leaf area could translate into more transpiration.
Rising carbon dioxide concentrations will decrease leaf stomatal conductance to water vapor. This effect could reduce transpiration.
Rising carbon dioxide concentrations and rising global temperatures could change WUE.

The following sections of this chapter will examine more closely the effects of rising carbon dioxide concentrations and climate change on vegetation, providing qualitative and quantitative assessments of how these changes will affect photosynthesis, growth, and transpiration water requirements of crops.

DIRECT EFFECTS OF CARBON DIOXIDE ON PHOTOSYNTHESIS, TRANSPIRATION, AND GROWTH OF PLANTS

Atmospheric Carbon Dioxide

The carbon dioxide concentration of the earth's atmosphere has varied throughout geologic time. Ice core data from Antarctica and Greenland have been obtained and, from entrapped air bubbles, used to show carbon dioxide and methane concentrations of the atmosphere throughout the past 160,000 years (Barnola et al., 1987; Lorius et al., 1990). Changes in the deuterium content within ice crystals have been used to establish temperature changes over this same time period (Jouzel et al., 1987). In general, carbon dioxide concentrations were as low as 180 to 200 parts per million (ppm) 13,000 to 30,000 years ago and 140,000 to 160,000 years ago during the coldest parts of the last two ice ages (Barnola et al., 1987). Carbon dioxide concentrations rose to about 270 ppm during the last interglacial period (116,000 to 140,000 years ago) and during the current interglacial period (beginning about 13,000 years ago). Ice core data since about 1700 A.D. and direct atmospheric sampling data since 1958 show that the carbon dioxide concentration increased to 315 ppm by 1958 and to about 355 ppm by 1990 (Keeling et al., 1989). The rate of increase of atmospheric carbon dioxide is about 0.5 percent per year, which means that the change is accelerating.

These changes in atmospheric carbon dioxide have important implications for plants and the global carbon cycle as well as for climate. Atmospheric carbon dioxide is the raw material for terrestrial green plant photosynthesis, and thus it represents the first molecular link in the food chain of the whole earth. In later sections, we will examine the importance of carbon dioxide for photosynthesis and plant growth, as well as the importance of potential climate change on water resources for the future.

Plant Photosynthetic Mechanisms

Three types of photosynthetic mechanisms of terrestrial green plants have been identified: C3, C4, and CAM. Responses of these three photosynthetic mechanisms to carbon dioxide have been reviewed by Tolbert and Zelitch (1983). The biochemical pathway of photosynthetic carbon dioxide uptake was first determined for C3 plant photosynthesis. This pathway involves the use and subsequent regeneration of ribulose 1,5-biophosphate in a cyclic series of reactions, and it is frequently called the Calvin cycle. The first product of photoassimilation of carbon dioxide is 3-phosphoglyceric acid, a three-carbon sugar—hence the term C3 pathway of photosynthesis.

The C4 plants begin their carbon dioxide uptake in a different process sometimes called the Hatch-Slack pathway. In mesophyll cells of leaves, these plants form a four-carbon molecule, oxalacetate, in the first step of incorporation of carbon dioxide. This four-carbon compound is changed into aspartic acid or malic acid and then transported immediately to bundle sheath cells. Here, the carbon dioxide is released and utilized in the C3 biochemical pathway. Thus, the C4 plant mechanism first traps carbon dioxide in the mesophyll cells, and then transports and concentrates the carbon dioxide in the bundle sheath cells, where it is utilized in C3 plant metabolism (Tolbert and Zelitch, 1983).

Crassulacean acid metabolism, or CAM, is a mechanism whereby plants typically take up and store carbon dioxide during the night and use it in photosynthetic carbon dioxide fixation during the day, when sunlight is available. Pineapple and ''air plants,'' such as Spanish moss and orchids, have this photosynthetic mechanism. Since few agricultural crops are CAM plants, they are not important in the process of managing water resources under conditions of climate uncertainty.

Since C4 plants have a mechanism for concentrating carbon dioxide in bundle sheath cells of leaves, their photosynthetic rates

will not respond to rising carbon dioxide levels to the same extent as C3 plants. Irrigated crop or turf plants that fit into the C4 category include maize (corn), sorghum, millet, sugar cane, and bermuda grass. Plants that fit into the C3 category include: wheat, rice, potato, soybean, sugar beet, alfalfa, cotton, tree and vine crops, and most vegetable crops and cool-season grasses.

Plant Growth Responses to Carbon Dioxide

Increasing atmospheric carbon dioxide levels have caused increasing photosynthetic rates, biomass growth, and seed yield for all of the globally important C3 food and feed crops (Acock and Allen, 1985; Enoch and Kimball, 1986; Warrick et al., 1986; Allen, 1990). Some plants, such as cucumber, cabbage, and perhaps tomato, have shown a tendency to first increase leaf photosynthetic rates in response to elevated carbon dioxide concentrations, and then to decrease photosynthetic rates after several days. This behavior is called "end-product inhibition of photosynthesis," and it is caused by the failure of translocation of photoassimilates to keep up with photosynthetic rates (Guinn and Mauney, 1980).

A few experiments have been conducted with carbon dioxide concentration maintained across a range of 160 to 990 ppm. Figure 7.3 shows the results of one study with soybean canopy photosynthetic rates across the 90 to 900 ppm carbon dioxide concentration range. A nonlinear hyperbolic model was used to fit soybean photosynthetic rate data to carbon dioxide concentration (Allen et al., 1987). Photosynthetic rates at the various carbon dioxide concentrations were divided by the photosynthetic rate at a carbon dioxide concentration of 330 ppm to normalize the data to a common condition. Data sets of biomass yield and seed yield from four locations over three years were also fit to the model (Allen et al., 1987). Relative yields with respect to yields at 330 to 340 ppm were used.

The form of the model fit to the experimental data was:

where

R = relative response of photosynthetic rate, biomass yield, or seed yield

R_max = asymptotic upper limit for R from baseline R_int

FIGURE 7.3 Photosynthetic carbon dioxide uptake rate responses of a soybean crop canopy exposed to carbon dioxide concentrations ranging from 110 to 990 ppm. All data points are relative to the response obtained at 330 ppm.

SOURCE: Adapted from Allen et al., 1987.

C = carbon dioxide concentration (ppm)

K_c = Apparent Michaelis constant (ppm)

R_int = Y-axis intercept for zero C

From the parameters of this equation, photosynthetic rate, biomass accumulation, and seed yield changes of soybean due to carbon dioxide concentration changes can be estimated (Allen et al.,

1987). Table 7.1 shows the changes predicted across three time periods: from the last ice age (when the carbon dioxide concentration was at a minimum) to the preindustrial revolution era (about 1700), from 1700 to 1973, and from 1973 to about a century into the future. The modeled data show that there should have been large increases in productivity between the ice-age (when carbon dioxide concentration was about 200 ppm) and the beginning of the industrial revolution (when the carbon dioxide concentration was about 270 ppm). Likewise, there should have been a 12 percent increase in grain-yield productivity between 1700 and 1973, when the carbon dioxide concentration increased from about 270 to 330 ppm.

Most of the recent concerns about rising atmospheric carbon dioxide concentrations have been quantified by predicting changes for a doubling of the carbon dioxide concentration, usually from 330 to 660 ppm. Table 7.1 shows that soybean seed yields and biomass yields are predicted to increase 31 percent and 41 percent, respectively, from a doubling of carbon dioxide. Experimental studies have consistently showed a lower seed yield than biomass yield for soybean when grown under a doubled carbon dioxide concentration. If the harvest index—the ratio of seed yield to above-ground biomass yield (seed plus pod walls plus stems)—were 0.50 for soybean grown under a 330 ppm carbon dioxide concentration, then the harvest index would be 0.46 if the carbon dioxide concentration were doubled. This small decrease in soybean harvest index under elevated carbon dioxide conditions has been commonly observed (Allen, 1990; Jones et al., 1984). The relative midday maximum photosynthetic rates under carbon dioxide enrichment were consistently higher than relative biomass yields, probably because the photosynthetic response to elevated carbon dioxide levels is greater under high light conditions than it is under total daily solar irradiance conditions.

Transpiration Responses to Carbon Dioxide

The effect of carbon dioxide concentration on water use under field conditions has been discussed for many years. In the past, elevated carbon dioxide levels have been mentioned as the ideal antitranspirant. This conclusion seems reasonable, since elevated carbon dioxide has been observed to reduce stomatal conductance in numerous experiments. Morison (1987) reviewed 80 observations in the literature and found that a doubled carbon dioxide con-

TABLE 7.1 Percent increases of soybean midday photosynthesis rates, biomass yield, and seed yield predicted across selected carbon dioxide concentration ranges associated with relevant benchmark points in time.

	CO₂ Concentration
Period of Time	Initial	Final	Midday Photosynthesis	Biomass Yield	Seed Yield
	--------ppm--------		------% increase over initial CO₂-----
IA–1700¹	200	270	38	33	24
1700–1973	270	330	19	16	12
1973–20??²	330	660	50	41	31
¹ IA, the Ice Age about 13,000 to 30,000 years before present. The atmospheric carbon dioxide concentrations that prevailed during the last Ice Age, and from the end of the glacial melt until preindustrial revolution times, were 200 and 270 ppm, respectively. ² The first world energy "crisis" occurred in 1973, when the carbon dioxide concentration was 330 ppm. This concentration is used as the basis for many carbon dioxide–doubling studies. The carbon dioxide concentration is expected to double sometime within the twenty–first century.

centration will reduce stomatal conductance of most plants by about 40 plus or minus 5 percent. Kimball and Idso (1983) calculated a 34 percent reduction in transpiration in response to a doubled carbon dioxide concentration in several short-term plant growth chamber experiments, which seems consistent with the review by Morison (1987). However, Morison and Gifford (1984) also showed that doubling carbon dioxide will cause a more rapid development of leaf area for many plants and hence an equal or greater transpiration rate in the early stages of plant growth, due to a more rapid development of transpiring surfaces. Therefore, increased rates of development of transpiring leaf surface offset the reduced stomatal conductance for water vapor.

Allen et al. (1985) and Allen (1990) also discussed the effect of reduction in stomatal conductance on foliage temperature. The cause-and-effect relationships can be summarized as follows: Any reduction in stomatal conductance due to increasing the carbon dioxide concentration will restrict transpiration rates per unit leaf area. A reduction in transpiration rates will result in less eva-

porative cooling of the leaves, and leaf temperatures will rise. As leaf temperatures rise, the vapor pressure inside the leaves will increase, and thus the leaf-to-air vapor pressure gradient, which is the driving force for transpiration, will increase. The increase in leaf vapor pressure will increase transpiration rates per unit leaf area; thus, the transpiration rates will be maintained at only slightly lower values than would exist at ambient environmental carbon dioxide concentrations. In effect, all of the energy balance factors involved in canopy foliage energy exchange—not just stomatal factors—must be considered.

Controlled environment studies of soybean at Gainesville, Florida, showed that canopy transpiration rate changes ranged from negative 2 percent (Jones et al., 1985a) to plus 11 percent (Jones et al., 1985b) for carbon dioxide treatments of 800 and 330 ppm with corresponding LAI values of 6.0 and 3.3. In another experiment in which differences in the LAI of soybean between the 330 ppm and the 660 ppm treatments were small (3.36 and 3.46, respectively), the seasonal cumulative water use decreased by 12 percent for the doubled carbon dioxide treatments (Jones et al., 1985c). Decreases were similar for both water-stressed and nonstressed treatments.

Field weighing lysimeter and neutron-probe water balance studies of cotton at Phoenix, Arizona, have shown evapotranspiration reductions due to elevated carbon dioxide levels ranging from 0 up to 9 percent (Kimball et al., 1983).

In conclusion, although stomatal conductance may be reduced by about 40 percent for doubled carbon dioxide concentrations, water use by C3 crop plants under field conditions will probably be reduced by only about 0 to 12 percent. If leaf area increases due to doubled carbon dioxide concentrations are small (or can be controlled), then the transpiration reductions may be meaningful, albeit small. If leaf area increases due to doubled carbon dioxide concentrations are large, then no reductions in transpiration are to be expected, and increases may be possible.

Streamflow Responses to Carbon Dioxide

Several attempts have been made to predict changes in streamflow due to an increase in carbon dioxide (Aston, 1984), changes in climate (Revelle and Waggoner, 1983), or both (Brazel and Idso, 1984). Aston (1984) modeled streamflow changes from a New South Wales, Australia, watershed over the course of a year based on reduction of stomatal conductance to one-half of current

values. His model predicted a 40 to 90 percent increase in annual streamflow above the observed baseline of about 150 mm per year from the actual watershed. However, Aston (1984) did not consider any increase of the LAI, which is perhaps a justifiable assumption for C4 plants but probably not for C3 plants. Rosenberg et al. (1990) conducted a quantitative analysis of evapotranspiration sensitivity to several plant and environmental factors. Their analysis demonstrated that increasing the LAI could indeed partially offset the effects of decreasing stomatal conductance on transpiration.

For climate change only, Revelle and Waggoner (1983) predicted that western river streamflows could be reduced by about 40 to 76 percent from the combined effects of a 2°C rise in temperature and a 10 percent reduction in precipitation. Brazel and Idso (1984) considered that vegetation would reduce transpiration to about two-thirds of its current value with a doubling of the carbon dioxide concentration, which led to predictions of increasing Arizona streamflow from about 63 to 460 percent. When they included a temperature increase of 2°C and a precipitation decrease of 10 percent, the predictions were still a 4 to 326 percent increase in streamflow. However, Brazel and Idso's predictions did not include any likely increases in vegetation LAI due to increased carbon dioxide levels. Although efforts to relate carbon dioxide and climate change impacts on water resources are continuing (Waggoner, 1990), realistic integration of vegetation influences on the hydrologic cycle are lacking.

Changes in vegetation may be a moot point when streamflow depends largely on spring snowmelt from lower elevations and continuous warm season snowmelt from higher elevations in the mountains of the West. The combination of complex plant responses and complex terrain make accurate hydrologic modeling a difficult task.

Plant Water-Use Efficiency

Allen et al. (1985) compared water-use efficiencies of soybean canopies grown in outdoor, sunlit, controlled-environment chambers at 800 and 330 ppm carbon dioxide concentrations which had LAI values of 6.0 and 3.3, respectively. For each of these treatments (two replications), the exposure carbon dioxide levels were cross-switched for one day. The ratio of the WUE values (i.e., WUE at 800 ppm carbon dioxide exposure divided by the WUE at 330 ppm

carbon dioxide exposure) averaged 2.33. The relative contributions of photosynthesis and transpiration to the ratio of WUE values were 73 and 27 percent, respectively. These comparisons are valid only for plant canopies with equal LAI values, because the same canopy was used for both carbon dioxide exposure levels. However, when the treatment and exposure levels of 800 ppm carbon dioxide were compared with the treatment and exposure levels of 330 ppm, the WUE ratio was 1.80, and the relative contributions to this ratio were 104 percent for photosynthesis and negative 4 percent for transpiration. The negative contribution of transpiration arises from the fact that canopy transpiration rates for the 800 ppm carbon dioxide treatment were slightly greater than the rates from the 330 ppm carbon dioxide treatment, due to the much larger LAI of the canopy exposed to the higher carbon dioxide treatment. Clearly, higher LAI values under elevated carbon dioxide concentrations can increase transpiration rates to the point where all of the improved WUE arises from increased photosynthetic rates and none from decreased water use.

Finally, it should be pointed out that increases in WUE in a world with higher carbon dioxide levels do not necessarily imply any reduction in crop water requirements per unit area of land. However, farmers should be able to achieve higher crop yields per unit land area with similar amounts of water. If temperatures rise, however, the overall WUE could actually decrease, because warmer climates have higher water requirements (as illustrated by Figure 7.2) and higher temperatures may cause yield reductions. The crop response scenarios that may affect hydrology and water resources management will be determined by the carbon dioxide and climate change scenarios and will differ depending on photosynthetic types (C4 versus C3) and species.

CLIMATE CHANGE EFFECTS ON PHOTOSYNTHESIS, GROWTH, AND TRANSPIRATION

Leaf photosynthetic rates are known to be sensitive to temperature. Figure 7.4 shows possible responses of leaf photosynthetic carbon dioxide uptake rates to temperature for C3 plants (bottom curve) and C4 plants (top curve) when grown at a 330 ppm carbon dioxide concentration and exposed to high light levels, such as would occur under midday summer conditions. This figure shows that C4 plants have a higher maximum photosynthetic carbon dioxide uptake rate and a higher temperature maximum than C3

FIGURE 7.4 Examples of maximum photosynthetic (PS) rate responses to temperature of individual leaves of C3 plants under high light conditions when exposed to carbon dioxide concentrations of 330 ppm (lower curve) and 1,000 ppm or greater (upper curve). The upper curve is similar to the maximum PS response to temperature of C4 plant leaves, which have an internal mechanism for concentrating carbon dioxide for subsequent photosynthetic reactions. Various species differ widely, both in maximum leaf photosynthetic rates and in the distribution of leaf photosynthetic rates with temperature.

SOURCE: Modified and adapted from the example of Pearcy and Björkman (1983). See also Berry and Björkman (1980) and Penning de Vries et al. (1989) for further examples of the variability of response among species and experimental conditions.

plants. The relative differences are smaller at lower temperatures. These curves were drawn to represent active crop plants in temperate zones. The actual photosynthetic carbon dioxide uptake rates

could be considerably different from those shown, and the temperature distribution of photosynthetic rates could be higher or lower, depending upon species, climate, or pretreatment temperature conditions (Berry and Björkman, 1980; Penning de Vries et al., 1989). In particular, the curves could be stretched to higher temperatures for species adapted to hot, desert environments (Pearcy and Björkman, 1983) or compressed to lower temperatures for species adapted to cool environments.

Nevertheless, from Figure 7.4 we can conclude that C4 plants could benefit more (or at least suffer less) than C3 plants from an increase in global temperatures. However, the differences for a whole canopy of leaves are somewhat reduced from the differences for individual leaves exposed perpendicularly to high light. First, a canopy of leaves generally has leaves oriented in all directions, so that much of the total leaf area is exposed to much less irradiance than in single leaf exposure systems. Under these conditions many of the individual leaves are limited by light, and the photosynthetic carbon dioxide uptake rates of the whole canopy become more similar. Second, solar irradiance levels are lower than midday values throughout much of the day. Nevertheless, the direction of the leaf-level differences, if not the magnitude, is maintained between C4 and C3 canopies.

Figure 7.4 also shows that C3 plant photosynthetic rates at elevated carbon dioxide levels may increase and resemble the rates of C4 plants (Pearcy and Björkman, 1983), but the extent of increase will vary widely among species. The photosynthetic rates of C3 plant leaves increase at elevated carbon dioxide levels because molecules of carbon dioxide compete more effectively with oxygen for binding sites on rubisco, the carboxylating enzyme (Bowes and Ogren, 1972).

When plants are well watered, leaf temperatures tend to rise more slowly than air temperatures throughout the daily cycle, so that foliage-to-air temperature differences become greater as air temperature rises (Idso et al., 1987; Allen, 1990). For soybean, Jones et al. (1985a) found no change in crop canopy photosynthetic rates across the air temperature set-point range of 28°C to 35°C. However, the transpiration rates increased 30 percent, which would lead to evaporative cooling of the leaves and larger foliage-to-air temperature differences. This 4 to 5 percent increase in transpiration rate per 1°C rise in temperature is close to the 6 percent per 1°C rise in saturation vapor pressure deficit over this temperature range.

Temperature affects growth of plants in several ways. The rate of development and expression of new nodes on plants increases

with increasing temperature. Leaves expressed at new nodes will grow larger, in general, if there is no concurrent water stress. Thus, plant size increases at a more rapid rate, and solar radiation capture occurs earlier in crop development. Once full ground cover is achieved, at a LAI of about 2 to 3, light capture becomes limiting, and the overall temperature effects on growth are muted but not eliminated. The duration of each ontogenic phase of plant growth decreases with increasing temperature, which is the most important effect of temperature within the upper and lower limits of survival.

INTERACTIVE EFFECTS OF CARBON DIOXIDE AND CLIMATE CHANGE

Photosynthetic and Productivity Interactions

As explained above, Figure 7.4 shows leaf photosynthetic carbon dioxide uptake rate versus temperature responses typical of C4 plants and C3 plants at carbon dioxide concentrations of 330 ppm. The upper curve can also represent C3 plants at high carbon dioxide levels of (1,000 ppm or greater). These curves suggest that a combination of rising carbon dioxide concentration and rising temperature should lead to greater photosynthetic rates and hence greater biomass growth rates.

S. G. Allen et al. (1988; 1990a,b) conducted experiments in Phoenix, Arizona, on Azolla, water lily, and sorghum as seasonal temperatures were changing. They found that net photosynthetic rates were higher for Azolla and water lily during warmer times of year. Linear regressions on net photosynthetic rates for water lily versus air temperature at the time the measurements were taken showed a greater increase with temperature for plants grown at a 640 ppm carbon dioxide concentration than for those grown at a 340 ppm carbon dioxide concentration. However, there was also an interaction with solar radiation. The plants grown at 640 ppm of carbon dioxide also showed a much greater response to solar radiation than those grown at 340 ppm. Although the interactions among carbon dioxide treatment level, air temperature, and solar radiation were not resolved, the data show that all were interrelated in the carbon dioxide response. S. G. Allen et al. (1990a,b) also computed linear regressions of photosynthetic rate versus previous minimum air temperatures and previous maximum air tem-

peratures for periods of 1, 3, 6, and 9 days. They found that net photosynthetic carbon dioxide uptake rates were more sensitive to previous minimum air temperatures than to previous maximum air temperatures. (Of course, maximum temperatures are closely related to minimum temperatures.) The slope of the regression increased with the number of previous days included in the calculation of air temperature. Most importantly, Allen et al., also found that leaf photosynthetic rates were more sensitive to previous minimum temperatures than to air temperature at the time the measurements were being made.

Photosynthetic rates of sorghum leaves increased only slightly when the leaves were grown at a high carbon dioxide concentration: the rates for a 640 ppm carbon dioxide concentration were about 10 percent higher than the rates for a 340 ppm temperature (S. G. Allen et al., 1990b). The effect of temperature on photosynthetic rates was also quite small—a response to be expected, since sorghum is a C4 plant. Leaves were about 1.0 to 1.5°C warmer under the 640 ppm carbon dioxide treatment.

Idso et al. (1987) compared growth rates (biomass accumulation rates) of carrot, radish, water hyacinth, Azolla, and cotton grown across a seasonal range of temperatures at carbon dioxide concentrations of 650 ppm with growth rates for these same crops grown at 350 ppm. They found that the ratio of biomass accumulation rates of plants grown at 650 ppm to plants grown at 350 ppm increased somewhat linearly with air temperature across the seasonal mean air temperature range of about 12°C to 36°C. The biomass growth ratio (biomass accumulation at specified elevated carbon dioxide concentration divided by biomass accumulation at a baseline carbon dioxide concentration) increased about 0.087 per 1°C over this temperature range and had a zero intercept at about 18.5°C mean daily air temperature. Under arid zone conditions at Phoenix, Arizona, daily minimum temperatures are 8 to 9°C lower than daily mean temperatures during the months of November to March, when mean air temperatures are well below 18.5°C. Low nocturnal temperatures, as well as low total solar irradiance, short photoperiod (day length), previous carbohydrate storage, and stage of growth of the plants may affect the biomass growth ratio during the winter months.

We may conclude that increasing both carbon dioxide concentrations and temperature will cause a greater increase in biomass productivity than increasing carbon dioxide levels alone. However, Baker et al. (1989) found different biomass growth ratios for both final harvest dry matter and seed yield for soybean grown at 330

and 660 ppm of carbon dioxide. Although early canopy vegetative growth rates suggest that the biomass growth ratio could increase with temperature, the final harvest data showed otherwise. The experiment was conducted with day/night air temperatures of 26/19, 31/24, and 36/29°C, which gave average air temperatures of about 22.8, 27.8, and 32.8°C under the 13/11 hour thermoperiod. The biomass growth ratios for final harvest dry matter were 1.50, 1.36, and 1.24 for the respective temperatures. The biomass growth ratios for final harvest seed yield were 1.46, 1.24, and 1.15 for the respective temperatures. The changes in the biomass growth ratio for dry matter and seed yield were -0.026 (r² = 0.98) and -0.031 (r² = 0.88) per 1°C, respectively. This cultivar of soybean, ''Bragg,'' has a determinate growth habit that causes vegetative growth to nearly cease when flowering begins. Furthermore, elevated temperatures tend to hasten maturity and shorten the life cycle of this soybean crop. These factors were different from the Arizona study. Furthermore, the study of Baker et al. (1989) had identical light conditions for all treatments, and photoperiod interaction with temperature was minimized by two weeks of supplemental lighting at the beginning of the season.

In summary, the biomass growth ratio of plants grown at elevated carbon dioxide concentrations may increase with increasing temperature for vegetative growth, as suggested by Figure 7.4. However, this response may be reversed for seed grain crops that have a determinate growth habit, such as "Bragg" soybean.

Evapotranspiration

Evapotranspiration refers to the combination of plant transpiration and evaporation directly from the soil surface. Much of the following discussion of evapotranspiration will refer largely to the effects of carbon dioxide and climate on the plant component, which is in general much larger than the soil component except when the LAI is small.

The best modeling studies to date on the simulated effects of climate change and carbon dioxide concentration increase on plant canopy evapotranspiration were conducted by Rosenberg et al. (1990), using the Penman-Monteith model. A similar approach was used by Allen and Gichuki (1989) and Allen et al. (1991) to estimate effects of carbon dioxide-induced climate changes on evapotranspiration and irrigation water requirements in the Great Plains from Texas to Nebraska. Rosenberg et al. (1990) examined the ef-

fects of temperature, net radiation, air vapor pressure, stomatal resistance, and LAI on three types of plant canopies: wheat at Mead, Nebraska; grassland at Konza Prairie, Kansas; and forest at Oak Ridge, Tennessee. Increasing the temperature by 3°C gave a 6 to 8 percent increase in transpiration per 1°C. This compares reasonably well with the 4 to 5 percent increase in transpiration per 1°C measured experimentally in soybean across the 28 to 35°C range by Jones et al. (1985a). Rosenberg et al. (1990) also reported that evapotranspiration decreased 12 to 17 percent for a 40 percent increase in stomatal resistance. This corresponds closely to a 12 percent decrease in seasonal transpiration obtained experimentally for soybean grown in controlled-environment chambers by Jones et al. (1985c) for doubled carbon dioxide concentration conditions when leaf area index was very similar for both the ambient and doubled carbon dioxide treatments. In another study, Jones et al. (1985b) showed that exposure of soybean canopies to a level of 800 ppm carbon dioxide decreased daily total transpiration by 16 percent in comparison to an exposure level of 330 ppm. Jones et al. attributed this reduction to an increase in stomatal resistance.

Increases in LAI of 15 percent caused increases in predicted evapotranspiration of about 5 to 7 percent according to the model of Rosenberg et al. (1990). These values were comparable to those extracted from Jones et al. (1985b). Their data showed a 33 percent increase in measured daily transpiration for a change in LAI from 3.3 to 6.0 (an 82 percent increase in LAI). However, the effect of LAI may not be linear. By use of a soil-plant-atmosphere model, Shawcraft et al. (1974) showed that the effect of LAI on transpiration would be highly nonlinear across a LAI range of 0 to 8. Most of the effect of changing leaf area occurred across the LAI range of 0 to 4. However, these simulations were conducted with a moist soil surface (having a water potential of -60 MPa) and relatively high soil surface-to-air boundary-layer conductance. Thus, predicted evapotranspiration rates at a LAI of 2 were maintained at 85 percent or more of the rates at a LAI of 8. Nevertheless, the modeling results of Shawcraft et al. (1974) for three solar elevation angles and three leaf elevation angle classes showed that predicted plant transpiration increased by an average of 27 plus or minus 8 percent for a LAI increase from 3.3 to 6.0.

Net radiation could increase under climate change conditions from both greater downwelling thermal radiation and increased solar radiation (decreased cloudiness), or it could decrease from increased cloudiness. Rosenberg et al. (1990) showed that evapotranspiration should change about 0.6 to 0.7 percent for each 1

percent change in net radiation. Likewise, they showed that evapotranspiration should change about -0.4 to -0.8 percent for each 1 percent change in vapor pressure of the air. A combination of several factors gave changes in evapotranspiration ranging from 27 percent (for a case of increased net radiation and decreased vapor pressure) to negative 4 percent (for a case of decreased net radiation and increased vapor pressure. These factors were: a temperature increase of 3°C, net radiation changes of plus or minus 10 percent, vapor pressure changes of plus or minus 10 percent, stomatal resistance increase of 40 percent, and leaf area index increase of 15 percent. Each factor related to climate change and plant response to carbon dioxide affects the predicted evapotranspiration.

ESTIMATING YIELD AND WATER REQUIREMENTS OF CROPS UNDER TWO CURRENT CLIMATE CHANGE SCENARIOS

General Circulation Models

Climate changes under conditions of doubled atmospheric carbon dioxide levels have been predicted using five atmospheric general circulation models (GCMs). These models are the National Oceanic and Atmospheric Administration (NOAA) Geophysical Fluids Dynamics Laboratory (GFDL) model developed at Princeton University (Manabe and Wetherald, 1986, 1987), the NASA Goddard Institute for Space Studies (GISS) model developed at Columbia University (Hansen et al., 1984, 1988), the Community Climate Models (CCM) developed at the National Center for Atmospheric Research (NCAR) (Washington and Meehl 1983, 1984, 1986), the Oregon State University (OSU) model (Schlesinger, 1984), and the United Kingdom Meterological Office (UKMO) model (Wilson and Mitchell, 1987; Mitchell, 1989). All of these models predict an increase of global average surface temperatures. The global mean surface temperature increases for recent modeling studies in which the carbon dioxide concentration was doubled were 2.8, 4.0, 4.0, 4.2 and 5.2°C for the OSU, CCM, GFDL, GISS, and UKMO models, respectively (Wilson and Mitchell, 1987). For a carbon dioxide concentration doubling, the global mean precipitation is also predicted to increase by 7.8, 7.1, 8.7, 11.0, and 15.0 percent for the above GCMs, respectively.

Grotch (1988) analyzed four of the GCM climate change scenarios for a climate with doubled carbon dioxide concentration.

Changes predicted across the United States were extracted and summarized, as were global, hemispherical, and other regional changes. The predicted June-July-August (JJA) median temperature increases for the United States were about 3.5, 3.0, 3.8, and 5.6°C for a carbon dioxide doubling from the OSU, CCM, GISS, and GFDL models, respectively. The predicted JJA changes in precipitation for the United States were 4, 10, 8, and -25 percent for the respective models. Thus, the GCM precipitation change scenarios are more variable for the United States than temperature change scenarios and may differ considerably among regions around the world. The UKMO model predictions for reduced precipitation for the United States are somewhat similar to the GFDL model scenario (Wilson and Mitchell, 1987). Higher JJA temperatures for the United States were associated with models with the lowest summer precipitation. As would be expected, both the GFDL model (Manabe and Wetherald, 1987) and the UKMO model (Wilson and Mitchell, 1987) predict serious decreases in soil wetness during the summer for the United States.

All of the GCMs predict a temperature increase for the Unites States for a doubling of atmospheric carbon dioxide. However, the predicted summer precipitation for the United States covers a range of 10 to -25 percent. The possibility of a significant reduction in summer precipitation, coupled with a temperature rise, could pose a serious problem for future agricultural productivity and water resources.

Modeling Crop Responses to Carbon Dioxide and Climate Changes

Many years of experimental observations on the interactions of carbon dioxide and climate factors would be required to provide complete information on responses of plants to climate change. However, plant growth models have been developed that are sensitive to environmental factors such as photoperiod, temperature, soil water availability, and light interception. These models can provide projections of crop response to future climate change scenarios in comparison with baseline climate records of the recent past.

Peart et al. (1989) and Curry et al. (1990a,b) used a soybean crop growth model, SOYGRO (Wilkerson et al., 1983; Jones et al., 1989), and a maize growth model, CERES-maize (Jones and Kiniry, 1986), for predicting growth and yield responses to doubled carbon dioxide climate change scenarios in the southeastern United States.

These simulations were conducted using 30 years of baseline weather data (1951 through 1980) from 19 sites in the southeastern United States—a region bounded by Virginia and Kentucky to the north and Arkansas and Louisiana to the west. Predicted climate changes within the appropriate grid cells of two GCMs, the GISS model (Hansen et al., 1988) and the GFDL model (Manabe and Wetherald, 1987), were used to change temperatures, precipitation, and solar radiation, month by month, for each of the baseline data sets at each site. These modified baseline data sets provided GISS and GFDL climatic scenarios (Smith and Tirpak, 1989). Monthly precipitation data of two sites for baseline climate and derived values for GISS and GFDL scenarios are given in Figures 7.5, 7.6, and 7.7, respectively, for Columbia, South Carolina, and Figures 7.8, 7.9, and 7.10, respectively, for Memphis, Tennessee. These two sites were essentially at the center of two GISS model grid cells and close to the center of two GFDL model grid cells, and thus should be appropriate sites for representing climate change scenarios within the grid cells of the two models. Monthly averages of July maximum daily temperatures for baseline, GISS, and GFDL scenarios were 33.27, 35.23, and 38.19°C, respectively, for Columbia, and 33.07, 35.44, and 36.03°C, respectively, for Memphis. Current planting dates, cultivars, and prevailing cropped soil types at each site were used in the simulations.

First, simulations were run based on rain-fed climate change conditions without direct fertilization effects of elevated carbon dioxide concentrations. Next, those simulations were repeated with optimum irrigation. Finally, simulations were run under both rain-fed and irrigated conditions for doubled atmospheric carbon dioxide conditions with a crop photosynthetic enhancement factor of 1.35 for soybean and 1.10 for maize.

Table 7.2 shows the soybean seed yield simulations from the SOYGRO model averaged over 19 sites and 30 years. First, under rain-fed conditions with climate change effects only, average soybean yield, compared to the baseline climate, was reduced by 71 percent under the GFDL scenario, but was reduced only 23 percent under the GISS scenario. The yields under the GFDL scenario were severely impacted because of the rainfall reductions predicted by this GCM (Figures 7.5 to 7.10).

Under optimum irrigation conditions, average soybean yields under both the GISS scenario and the GFDL scenario were reduced by 18 or 19 percent with respect to optimum irrigation under baseline climate conditions. However, in spite of higher temperatures, the irrigated yields under the GISS and GFDL scenarios were about

FIGURE 7.5 Average monthly precipitation and potential evapotranspiration for Columbia, South Carolina, for the 30-year base climate period, 1951 to 1980.

SOURCE: Precipitation data adapted from Table 6 of Peart et al., 1989.

FIGURE 7.6 Derived average monthly precipitation and potential evapotranspiration for Columbia, South Carolina, from a GISS climate change scenario for doubled atmospheric carbon dioxide.

SOURCE: Precipitation values adapted from Table 6 of Peart et al., 1989.

FIGURE 7.7 Derived average monthly precipitation and potential evapotranspiration for Columbia, South Carolina, from a GFDL climate change scenario for doubled atmospheric carbon dioxide.

SOURCE: Precipitation values adapted from Table 6 of Peart et al., 1989.

FIGURE 7.8 Average monthly precipitation and potential evapotranspiration for Memphis, Tennessee, for the 30-year base climate period, 1951 to 1980.

SOURCE: Precipitation data adapted from Table 6 of Peart et al., 1989.

FIGURE 7.9 Derived average monthly precipitation and potential evapotranspiration for Memphis, Tennessee from a GISS climate change scenario for doubled atmospheric carbon dioxide.

SOURCE: Precipitation data values from Table 6 of Peart et al., 1989.

FIGURE 7.10 Derived average monthly precipitation and potential evapotranspiration for Memphis, Tennessee from a GFDL climate change scenario for doubled atmospheric carbon dioxide.

SOURCE: Precipitation values adapted from Table 6 of Peart et al., 1989.

TABLE 7.2 Effects of climate change only on simulated soybean yields (bushels/acre) for the southeastern United States when the atmospheric carbon dioxide concentration is doubled.

BASE	GISS Model		GFDL Model		Model
Yield	Yield	Diff.	Yield	Diff.	Diff.
------------------------------------------Rainfed------------------------------------------
37	29	-23%	11	-71%	62%
------------------------------------------Irrigated-------------------------------------------
57	47	-18%	46	-19%	-2%
SOURCE: Adapted from Peart et al., 1989.

25 percent greater than the baseline climate scenario without irrigation.

When carbon dioxide fertilization plus climate change effects were simulated (Table 7.3), average soybean yields for the GISS climate scenario increased 11 percent under rain-fed conditions, whereas yields under the GFDL climate scenario still decreased (by 52 percent). Under optimum irrigation with carbon dioxide fertilization, yields under both GISS and GFDL scenarios were increased 13 to 14 percent relative to the irrigated baseline climate scenario.

Maize yields declined by only about 8 percent in the GISS climate scenario and by about 73 percent in the GFDL scenario when the effects of climate change due to the greenhouse effect were simulated (Table 7.4). Although irrigation increased predicted yields, the GISS and GFDL climate scenarios gave yield decreases of 18 and 27 percent, respectively, relative to the irrigated baseline weather crops. Including the direct effects of carbon dioxide fertilization plus climate change had little effect on the predicted yields of maize (Table 7.5), as expected because maize is a C4 plant.

Ritchie et al. (1989) found that of the climate changes that might occur because of increasing greenhouse gas concentrations, higher temperature had the greatest effect on predicted soybean and maize yields in the Great Lakes and corn belt regions. Increases in temperature caused a decrease in the duration of the crop life cycle. Yield reductions were greatest for the GFDL

TABLE 7.3 Direct carbon dioxide fertilization effects plus climate change effects on simulated soybean yields (bushels/acre) for the southeastern United States under a doubled carbon dioxide concentration.

BASE	GISS Model		GFDL Model		Model
Yield	Yield	Diff.	Yield	Diff.	Diff.
------------------------------------------Rainfed------------------------------------------
37	41	+11%	18	-52%	-56%
------------------------------------------Irrigated------------------------------------------
57	65	-13%	65	-14%	+1%
SOURCE: Adapted from Peart et al., 1989.

TABLE 7.4 The effects of climate change only on simulated maize yields (bushels/acre) when the atmospheric carbon dioxide concentration is doubled. Effects are averaged over four locations: Charlotte, Macon, Meridian, and Memphis.

BASE	GISS Model		GFDL Model		Model
Yield	Yield	Diff.	Yield	Diff.	Diff.
------------------------------------------Rainfed------------------------------------------
137	126	-8%	37	-73%	-71%
------------------------------------------Irrigated------------------------------------------
224	183	-18%	164	-27%	-10%
SOURCE: Adapted from Peart et al., 1989.

climate change scenario for the southernmost locations of this region. Predicted yields increased for the northernmost stations because temperatures and growing season duration became more favorable for these crops. Overall, irrigation water requirements

TABLE 7.5 Direct carbon dioxide fertilization effects plus climate change effects (for a doubled carbon dioxide concentration) on simulated maize yields (bushels/acre) averaged over four locations: Charlotte, Macon, Meridian, and Memphis.

BASE	GISS Model		GFDL Model		Model
Yield	Yield	Diff.	Yield	Diff.	Diff.
------------------------------------------Rainfed------------------------------------------
137	130	-5%	35	-74%	-73%
------------------------------------------Irrigated------------------------------------------
224	184	-18%	165	-26%	-10%
SOURCE: Adapted from Peart et al., 1989.

in this region increased about 90 percent for the GFDL scenario and decreased about 30 percent for the GISS scenario.

Rosenzweig (1989) modeled corn and wheat yields in the Great Plains under GISS and GFDL climate change scenarios. She found that corn and wheat yields decreases were most extreme in the southern Great Plains because higher temperatures shortened the life cycle of the crops. Where precipitation was predicted to decrease, irrigation requirements increased. Allen and Gichuki (1989) predicted a 15 percent overall irrigation requirement increase for this region, with greater requirements for alfalfa because its growing season was increased and lower requirements for corn and winter wheat because their growing seasons were decreased.

The direct effect of rising carbon dioxide concentrations offset the adverse effects of climate change at some, but not all, locations in the simulations of Ritchie et al. (1989) and Rosenzweig (1989).

The impact of climate change on California water resources and irrigated agriculture is of particular interest because of the wide range of vegetable, fruit, and nut crops produced there for the rest of the nation. Dudek (1989) predicted productivity changes for several of these crops in response to GISS and GFDL climate change scenarios in which the carbon dioxide concentration was doubled; the basis for Dudek's work was a United Nations Food and Agriculture Organization agro-ecological zone method

(Doorenbos and Kassam, 1979). This method is not as rigorous as crop climate models, but such models have not been developed for specialty crops. For climate change only, Dudek (1989) predicted productivity decreases of 3 to 40 percent, depending upon crop, region, and scenario. Depending upon crop, statewide average yield decreased by 8 to 34 percent for the GISS scenario and by 6 to 31 percent for the GFDL scenario. When carbon dioxide enrichment effects were introduced, based in part on information from Kimball (1983), predicted productivity changes ranged from plus 41 to minus 27 percent, depending upon crop, region, and scenario. However, ranges of statewide average yield changes were somewhat smaller. Depending upon crop, predicted yields ranged from about plus 17 to minus 12 percent for the GISS scenario and about plus 21 to minus 8 percent for the GFDL scenario.

Dudek (1989) used the California agriculture and resource model (CARM) to predict economic and market impacts of the productivity changes. Although production changes generally declined under the climate change only scenarios, commodity prices generally increased. Under the climate change plus carbon dioxide effects scenarios, the overall impacts were much smaller. Since predicted water resource supplies were reduced, especially for the San Joaquin Valley, predicted use of both ground water and surface water decreased.

Rosenberg et al. (1990) used the Penman-Monteith evapotranspiration model and GISS, GFDL, and NCAR scenarios for estimating the impact of predicted climate changes on summer evapotranspiration at Mead, Nebraska (for wheat), Konza Prairie, Kansas (for grassland), and Oak Ridge, Tennessee (for forest). They found that inclusion of other factors (net radiation, vapor pressure, wind speed, stomatal resistance, and leaf area index) reduced the impact of temperature increases on predicted evapotranspiration. For example, a 6.3°C rise in GFDL scenario temperature for Mead, Nebraska, resulted in a 42 percent increase in predicted evapotranspiration but only a 23 percent increase when all other climate change factors were also considered. Similarly, a GISS scenario temperature rise of 4.7°C for Konza Prairie, Kansas, gave a 28 percent predicted increase in evapotranspiration, but this increase was only 4 percent when all other climate change factors were included in the Penman-Monteith equation. On the basis of these computations, Rosenberg et al. (1990) caution against the presumption that evapotranspiration will increase in all scenarios of global climate change. When the decreased duration of the life cycle of plants (caused by increased temperature) is included in climate change

scenarios, some crops (such as corn and winter wheat) may actually transpire less total water throughout their life cycles when grown on the Great Plains, as shown in model studies of Allen and Gichuki (1989) and Allen et al. (1991).

IMPLICATIONS OF CHANGING EVAPOTRANSPIRATION AND PRECIPITATION FOR WATER RESOURCES

Adaptations and Evapotranspiration Requirements

The crop yield simulations of the previous section demonstrate two main points: the adverse impact on crop production of inadequate rainfall under rising temperature scenarios and the importance of direct carbon dioxide effects in the face of elevated temperatures. However, the climate changes that would occur under the doubled carbon dioxide concentrations used in this simulation may occur at lower carbon dioxide concentrations if radiatively active trace gases other than carbon dioxide play a large role in the greenhouse effect. In that case, the direct carbon dioxide effects would be somewhat lower than those shown in the examples of Tables 7.2 through 7.5 for an equivalent climate change.

Much of the reduction in soybean yields reported by Peart et al. (1989) and Curry et al. (1990a,b) was due to decreases in the length of the grain filling period under higher temperatures. Changes in management practices may help restore part of the yield. Planting earlier or later in the season may help offset the higher temperatures. Selection of other cultivars could also help. In the future, it may be necessary to breed plants with new combinations of temperature tolerance and photoperiod responses, or perhaps to use growth regulators. Under conditions in which nonstructural carbohydrates accumulate in carbon dioxide-enriched plants, new germplasms that use photoassimilate more efficiently need to be developed.

Irrigation is not likely to be a panacea for climate change. In simulations by Peart et al. (1989) and Curry et al. (1990a,b), the average irrigation requirement increased by 33 percent under the GISS scenario and 134 percent under the GFDL scenario. However, with less summertime rainfall under the GFDL scenario (Figures 7.5 through 7.10), region-wide water resources would become scarce, and water may not be readily available for crops. Some areas of the United States may have to adapt by irrigating less land area and other areas by shifting from rain-fed to irrigated agriculture.

Increasing temperatures and decreasing precipitation for the nation as predicted by the GFDL model (and the UKMO model) would have serious negative overall impacts on agricultural productivity and society, although producers in some regions may benefit from higher prices brought on by scarcity (Adams et al., 1990).

Water Availability

Changes in crop yields and irrigation requirements caused by changed rainfall patterns and temperatures would influence ground water recharge, streamflow, and reservoir storage, as illustrated by Miller and Brock (1988; 1989) in a Tennessee Valley Authority modeling study of climate change. Climate change scenarios for Columbia, South Carolina, and Memphis, Tennessee for the GISS and GFDL models were selected to further illustrate the precipitation and temperature change effects on potential monthly and annual evapotranspiration and water deficits or excesses (Peart et al., 1989; Curry et al., 1990a,b).

Monthly potential evapotranspiration (PET) was calculated from the method of Jensen and Haise (1963) as modified by Jensen (1966). Cloud cover data were taken from Landers (1974) for Columbia and from Dickson (1974) for Memphis. Monthly solar radiation data were calculated from extraterrestrial solar irradiance and cloud cover data according to the procedure shown by Doorenbos and Pruitt (1977). The monthly Jensen-Haise PET computations were conducted based on computer programs by Zazueta and Smajstrla (1989). The summation of these monthly PET values was about 20 percent larger than values shown by Geraghty et al. (1973) in a water atlas of the United States. The method of Stephens and Stewart (1963), with coefficients applicable for Waynesville, North Carolina (Stephens, 1965), was applied to the Memphis baseline (1951 through 1980) climate data. Hand calculations using the Stephens-Stewart method gave an annual PET of 895 mm for Memphis, which was 84 percent of the computed Jensen-Haise value and was in close agreement with the water atlas value (Geraghty et al., 1973). Therefore, all monthly Jensen-Haise PET values were adjusted by multiplying them by 0.84, which gave annual PET values of 895 mm for Memphis and 955 mm for Columbia.

The average annual temperatures for the baseline period, the GISS scenario, and the GFDL scenario were 17.4, 19.9, and 20.6°C, respectively, for Columbia and 16.5, 19.9, and 19.3°C, respectively,

for Memphis. Average monthly distributions of calculated PET, along with precipitation (PPT), for Columbia are shown in Figures 7.5, 7.6, and 7.7 for the baseline climate, the GISS scenario, and the GFDL scenario, respectively. On average, monthly PPT for the May through October period would be sufficient for crop production for both baseline climate and GISS scenario conditions (Table 7.6). However, PPT minus PET over the May through October period was -435 mm for the GFDL scenario, which indicates a severe water deficit during this period. Nevertheless, during the November through April period, PPT minus PET for the GFDL scenario was about 87 percent of the GISS scenario. Therefore, following dry summer months there is opportunity for sizable recharge and streamflow during the winter months despite the overall annual water deficit of about 153 cm for the GFDL scenario at Columbia (Table 7.6).

For the Memphis example, the baseline climate showed a slight tendency toward summer drought (Figure 7.8). The GISS scenario (Figure 7.9) and the GFDL scenario (Figure 7.10) showed more month-to-month variation for Memphis than for Columbia. However, the total annual PPT was similar for all three Memphis cases (Table 7.6). The PPT deficit was particularly severe for the GFDL scenario during the June through August period. Nevertheless, the potential for recharge and subsequent streamflow was great (528 mm) during the November through April period, so the GFDL scenario has the potential of providing more streamflow water resources than the GISS scenario, although the GISS scenario provides more rainfall for crops and other vegetation during the summer months (Table 7.6).

These examples at two locations in a humid region show that although climate change scenarios may imply severe rainfall shortages and soil water deficits during the growing period, PPT excesses over PET during the cooler parts of the year still have the potential for maintaining cool season streamflows and water storage in reservoirs.

Allen et al. (1982) showed that annual streamflow water yield of a Florida watershed was linearly related to annual PPT. Furthermore, the intercept on the PPT axis was 806 mm, which approached the average annual watershed evapotranspiration of 891 mm (Knisel et al., 1985) determined by water budget analysis. Knisel et al. (1985) further showed that streamflow in this watershed was generally linearly related to PPT during three 4-month periods over the annual cycle. The intercept on the PPT axis for each of the three periods was similar to the average

TABLE 7.6 Average annual precipitation (PPT), potential evapotranspiration (PET), PPT minus PET for a 6-month period of recharge and runoff (November through April), and PPT for a 6-month period of high PET (May through October) for baseline climate (1951 to 1980) and for GISS and GFDL scenarios at two locations: Columbia and Memphis.

	BASE (mm)	GISS (mm)	GFDL (mm)
Columbia, SC
PPT	1245	1419	1042
PET	955	1096	1195
PPT-PET (Nov.-Apr.)	+317	+323	+282
PPT-PET (May-Oct.)	-27	0	-435
NET	+290	+323	-153
Memphis, TN
PPT	1310	1314	1306
PET	895	1063	1061
PPT-PET (Nov.-Apr.)	+522	+361	+528
PPT-PET (May-Oct.)	-107	-110	-283
NET	+415	+251	+245

watershed evapotranspiration during each period. These studies demonstrated that, in general, evapotranspiration requirements had to be met before sizable amounts of runoff occurred. However, intense or long-lasting rainfall events and low soil infiltration and percolation would result in increased runoff and streamflow before the monthly PET requirements are met. Also, during cooler months with higher rainfall, some soil water recharge must occur before rainfall will produce streamflow.

The impact of several climate change scenarios on California water resources was summarized by King et al. (1989). In general,

the GISS and GFDL models predicted annual temperature increases of about 4.6 and 4.4°C, respectively, whereas the OSU model predicted temperature increases of only 2.1°C. These predicted GISS and GFDL temperature changes would lead to greater winter runoff from snowmelt and rainfall in the mountains around the Central Valley, and the snowpack would melt earlier. Consequently, runoff in the late spring and early summer would be less. The predicted temperature changes of the OSU model led to only slight changes in runoff patterns, with peak runoff still occurring in May. The current reservoir storage and subsequent deliveries of the California State Water Project system could be reduced by 16, 14, and 7 percent based on the GISS, GFDL, and OSU climate change scenarios (King et al., 1989). According to the projections of the climate change scenarios, the Federal Central Valley Project would not be impacted as much.

Several assessments of climate change on water resources throughout the United States have been reported (e.g., Smith and Tirpak, 1989; Gleick, 1990; Waggoner, 1990). In general, watersheds in the West and Great Plains (except for those in the Pacific Northwest) have a high consumptive use to renewable supply ratio (averaging about 0.25) and a high total demand (the sum of consumptive use, ground water overdraft, water transfers, and evaporation) to renewable supply ratio (averaging about 0.35). Also, except for in the Pacific Northwest and the upper Colorado River basin, the ratio of ground water overdraft to ground water recharge is large (averaging about 0.42). These factors illustrate the vulnerability of the western states' water resources to climate change scenarios that include reduced precipitation. The impacts of reduced precipitation on native ecosystems, rain-fed agriculture, and water supply systems could be equally severe for eastern states. Variability of precipitation causes water shortages under present conditions, as demonstrated by the current California drought. Therefore, not only climate changes but also climate variability must be considered in managing water resources for sustained agricultural production to ensure an adequate food supply. Studies of interannual and daily variability of temperatures and precipitation show variations, in turn, among GCM models and among climate scenarios driven by both current and future projected levels of greenhouse gases (Mearns, 1989). Whether we should anticipate the consequences of a hotter and drier climate or a hotter and wetter climate, with more, less, or no change in variability, remains to be seen.

SUMMARY

Elevated carbon dioxide levels appear to increase the size and dry weight of most plants and plant components. Relatively more photoassimilate appears to be partitioned into structural components (stems and petioles) during vegetative development in order to support the light-harvesting apparatus (leaves). This observation may be a manifestation of plant size rather than a unique carbon dioxide effect. In general, the harvest index tends to decrease with increasing carbon dioxide concentration and with increasing temperature. Selection of plants that would use more photoassimilates for reproductive growth seems a useful goal for future research as the atmospheric carbon dioxide concentration increases. Plant growth regulators may also play an important role. Research efforts should be directed not only toward assessing the impacts of climate change on agriculture but also toward exploring biological adaptations and management systems for reducing the impacts of climate change. Whether we should anticipate a hotter and drier climate or a hotter and wetter climate remains to be seen.

The key topics discussed in this paper are summarized below:

The biomass and seed yield production of crops is linearly related to water use. The ratio of dry matter produced (measured as, for example, total biomass, or seed yield) to crop water used is called water-use efficiency (WUE). More water is required per unit biomass produced (in other words, the WUE is lower) in hot, arid environments than in cool, humid environments (which have a higher WUE). The water requirement also varies with species. Under current carbon dioxide levels, plants with the C4 photosynthetic mechanism require less water per unit biomass produced (they have a higher WUE) than plants with the C3 photosynthetic mechanism (which have a lower WUE).
The carbon dioxide concentration was as low as 180 to 200 ppm during the coldest part of the last ice age. As the ice melted, the concentration rose to about 270 ppm, where it remained stable until the advent of the industrial revolution and expansion of human population. The carbon dioxide concentration increased to 315 ppm by 1958, when continuous measurements were begun at Mauna Loa, Hawaii. The atmospheric carbon dioxide concentration rose to 330 ppm by 1973, the year of the first ''energy crisis,'' and today has reached about 355 ppm. Based on soybean experiments, the increase from 270 to 330 ppm should have increased seed yields by 12 percent. If carbon dioxide levels double from 330 to 660

ppm, seed yield should increase by about 31 percent, which is in agreement with a 33 plus or minus 6 percent increase in plant growth and yields reported in a literature survey by Kimball (1983).

Doubling the carbon dioxide concentration will cause stomatal conductance of leaves to decrease by about 40 percent. However, whole crop cumulative transpiration over the course of a season may only be reduced by 0 to 12 percent, for two reasons. First, when stomata partially close and restrict vapor from leaves, the foliage temperature rises. This raises the vapor pressure of water in the intercellular air space of leaves and increases the leaf-to-air vapor pressure difference. Thus, the reduction of stomatal conductance is substantially overcome by the driving force for evaporation. Secondly, elevated carbon dioxide levels promote growth of a greater amount of leaf area, so that a larger surface area for transpiration exists. In conclusion, a 40 percent reduction in stomatal conductance probably provides only a 10 to 15 percent reduction in transpirational water use. The greater amount of leaf area under elevated carbon dioxide conditions could eliminate this small difference in crop water use.
Increases in WUE of carbon dioxide-enriched crops is due largely to sizable increases in photosynthesis, growth, and yield. Decreases in water use are small and contribute very little, if anything, to increases in WUE.
Plants exposed to elevated carbon dioxide concentrations (650 ppm in comparison with 350 ppm) have shown greater response to carbon dioxide at high average daily temperatures than at low average daily temperatures during vegetative growth across the range of 12 to 35°C. This response is in agreement with single leaf photosynthetic measurements of responses to elevated carbon dioxide across a range of temperatures. One could conclude that all crops will respond more to the combination of increasing temperature with elevated carbon dioxide than to elevated carbon dioxide levels alone. However, during reproductive growth of soybean plants, the opposite trend was found. The complex responses of various kinds of plants to interactions of carbon dioxide, temperature, water supply, light, and photoperiod (day length) need further research.
Increasing temperatures across the range of 28 to 35°C appears to increase the transpiration rate by about 4 to 5 percent per 1°C, as shown in both experimental and modeling studies. This is in close agreement with the rise in saturation vapor pressure of about 6 percent per 1°C.

Outputs from two GCMs, the GISS and GFDL models, for a doubled carbon dioxide concentration were used as examples of predictions of soybean crop yield in the southeastern United States. The GISS model predicted about 25 percent more June through August rainfall than baseline 1951 through 1980 values, and the GFDL model predicted about 40 percent less rainfall. Under rainfed conditions and considering only the effects of climate change, the soybean crop model predicted a decrease in yields of 23 and 71 percent for the GISS and GFDL scenarios, respectively, compared with baseline weather conditions. When the direct effect of carbon dioxide on plant growth was included, the predictions were plus 11 percent for the GISS model and minus 52 percent for the GFDL model. When the simulated crops were irrigated, there was little difference between the predicted yields (plus 13 to 14 percent) of the two climate change scenarios. These data illustrate the critical importance of temperature and especially rainfall in any climate change scenario. Predicted crop yields for other regions (the Great Lakes, corn belt, Great Plains, and California) in response to carbon dioxide and climate change were generally similar to those of the Southeast.
Under the soybean simulations for the southeastern United States, the average requirement for irrigation increased by 33 percent and 134 percent above the baseline for the GISS and GFDL scenarios, respectively.
Rising carbon dioxide levels will cause an increase of photosynthetic rates, growth, yield, and water use efficiency for C3 crop plants. Water use per unit land area will not change much unless temperatures increase. Elevated temperatures may reduce seed crop yields in most areas of production, and potential rainfall decreases could cause serious reductions in rain-fed agriculture. Under irrigated agriculture conditions, reductions of precipitation could limit the number of acres available for irrigation and could lead to serious competition for water resources among various users.
Changes in precipitation amounts and distribution are the most serious climate change factors, from the standpoint of both crop productivity and water resources. However, low rainfall during the summer growing season does not necessarily mean a large reduction in overall water yield of a basin. If high rainfall should occur during the cool seasons, annual runoff, streamflow, and reservoir storage may not be impacted as much as if there were no change in seasonal patterns. Thus, water resources for nonagricultural uses may not be impacted as much as water for agriculture. Models that incorporate sufficient details of the hydro-

logic cycle, as well as vegetation and energy balance factors, should be developed to provide a more informed physical basis for managing water resources. This information could be used in assessing long-range ecological, food supply, economic, and societal consequences of water management decisions under conditions of climate uncertainty.

ACKNOWLEDGMENTS

This work was supported in part by the U.S. Department of Energy Interagency Agreements DE-AI05-88ER69014 and DE-AI01-81ER60001 with the U.S. Department of Agriculture, Agricultural Research Service. This work was conducted in cooperation with the University of Florida at Gainesville. Florida Agricultural Experiment Station Journal Series number R-01423.

REFERENCES

Acock, B., and L. H. Allen, Jr. 1985. Crop responses to elevated carbon dioxide concentration. Pp. 53-97 in B. R. Strain and J. D. Cure, eds., Direct Effects of Increasing Carbon Dioxide on Vegetation. Report DOE/ER-0238. Washington, D.C.: U.S. Department of Energy, Carbon Dioxide Research Division.

Adams, R. M., C. Rosenzweig, R. M. Peart, J. T. Ritchie, B. A. McCarl, J. D. Glyer, R. B. Curry, J. W. Jones, K. J. Boote, and L. H. Allen, Jr. 1990. Global climate change and U.S. agriculture. Nature 345:219-224.

Allen, L. H., Jr. 1990. Plant responses to rising carbon dioxide and potential interactions with air pollutants. J. Environ. Qual. 19:15-34.

Allen, L. H., Jr., K. J. Boote, J. W. Jones, P. H. Jones, R. R. Valle, B. Acock, H. H. Rogers, and R. C. Dahlman. 1987. Response of vegetation to rising carbon dioxide: Photosynthesis, biomass, and seed yield of soybean. Global Biogeochemical Cycles 1:1-14.

Allen, L. H., Jr., P. Jones, and J. W. Jones. 1985. Rising atmospheric CO₂ and evapotranspiration. Pp. 13-27 in Advances in Evapotranspiration. ASAE Pub. 14-85. St. Joseph, Michigan: American Society of Agricultural Engineers.

Allen, L. H., Jr., W. G. Knisel, and P. Yates. 1982. Evapotranspiration, rainfall, and water yield in south Florida research watersheds. Soil Crop Sci. Soc. Fla. Proc. 41:127-139.

Allen, R. G., and F. N. Gichuki. 1989. Effects of projected CO₂-induced climatic changes on irrigation water requirements in the Great Plains States (Texas, Oklahoma, Kansas, and Nebraska). Appendix C (Agriculture), Vol. 1, Chapter 6 in J. B. Smith and D. A. Tirpak, eds., The Potential Effects of Global Climate Change on the United States. Report EPA-230-05-89-053. Washington, D.C.: U.S. Environmental Protection Agency.

Allen, R. G., F. N. Gichuki, and C. Rosenzweig. 1991. CO₂-induced climatic changes on irrigation water requirements. J. Water Resources Planning and Management 117:157-178.

Allen, S. G., S. B. Idso, B. A. Kimball, and M. G. Anderson. 1988. Interactive effects of CO₂ and environment on photosynthesis of Azolla . Agric. For. Meteorol. 42:209-217.

Allen, S. G., S. B. Idso, and B. A. Kimball. 1990a. Interactive effects of CO₂ and environment on net photosynthesis of water lily. Agric., Ecosystems, and Environ. 30:81-88.

Allen, S. G., S. B. Idso, B. A. Kimball, J. T. Baker, L. H. Allen, Jr., J. R. Mauney, J. W. Radin, and M. G. Anderson. 1990b. Effects of Air Temperature on Atmospheric CO₂-Plant Growth Relationships--TR048 . Report DOE/ER-0450T. Washington, D.C.: U.S. Department of Energy and U.S. Department of Agriculture.

Allison, F.E., E. M. Roller, and W. A. Raney. 1958. Relationship between evapotranspiration and yield of crops grown in lysimeters receiving natural rainfall. Agron. J. 50:506-511.

Arkley, R. J. 1963. Relationship between plant growth and transpiration. Hilgardia 34:559-584.

Aston, A. R. 1984. The effect of doubling atmospheric CO₂ on streamflow: a simulation. Jour. Hydrol. 67:273-280.

Baker, J. T., L. H. Allen, Jr., K. J. Boote, P. Jones, and J. W. Jones. 1989. Response of soybean to air temperature and carbon dioxide concentration. Crop Sci. 29:98-105.

Barnola, J. M., D. Raynaud, Y. S. Korotkevich, and C. Lorius. 1987. Vostok ice core provides 160,000-year record of atmospheric CO₂. Nature 329:408-414.

Bennett, O. I., B. D. Doss, D. A. Ashley, V. J. Kilmer, and E. C. Richardson. 1964. Effects of soil moisture regime on yield, nutrient content, and evapotranspiration for three annual forage species. Agron. J. 56:195-198.

Berry, J., and O. Björkman. 1980. Photosynthetic response and adaptation to temperature in higher plants. Ann. Rev. Plant Physiol. 31:491-543.

Bowes, G., and W. L. Ogren. 1972. Oxygen inhibition and other properties of soybean ribulose-1, 5-diphosphate carboxylase. J. Biol. Chem. 247:2171-2176.

Brazel, A. J., and S. B. Idso. 1984. Implications of the Rapidly Rising CO₂ Content of the Earth's Atmosphere for Water Resources in Arizona. Scientific Paper No. 19, Laboratory of Climatology. Tempe, Arizona: Arizona State University.

Briggs, L. J., and H. L. Shantz. 1913a. The Water Requirements of Plants: I. Investigations in the Great Plains in 1910 and 1911. Bureau of Plant Industry Bull. 284. Washington, D.C.: U.S. Department of Agriculture.

Briggs, L. J., and H. L. Shantz. 1913b. The Water Requirements of Plants: II. A Review of the Literature. Bureau of Plant Industry Bull. 285. Washington, D.C.: U.S. Department of Agriculture.

Briggs, L. J., and H. L. Shantz. 1914. Relative water requirements of plants. J. Agric. Res. 3:1-63.

Chang, J. H. 1968. Climate and Agriculture, An Ecological Survey. Chicago: Aldine Publishing Co.

Curry, R. B., R. M. Peart, J. W. Jones, K. J. Boote, and L. H. Allen, Jr. 1990a. Simulation as a tool for analyzing crop response to climate change. Trans. ASAE 33:981-990.

Curry, R. B., R. M. Peart, J. W. Jones, K. J. Boote, and L. H. Allen, Jr. 1990b. Response of crop yield to predicted changes in climate and atmospheric CO₂ using simulation. Trans. ASAE 33:1381-1390.

De Wit, C. T. 1958. Transpiration and Crop Yields. Versl. Landbouwk. Onderz. (Agr. Res. Rep.) 64.6. Wageningen, Netherlands: Centre for Agricultural Publication and Documentation (Pudoc).

Dickson, R. R. 1974. The climate of Tennessee. Pp. 370-384 in Climate of the States, Vol. 1. Port Washington, New York: Water Information Center, Inc.

Doorenbos, J., and A. H. Kassam. 1979. Yield Responses to Water. FAO Irrigation and Drainage Paper No. 33. Rome, Italy: Food and Agriculture Organization of the United Nations.

Doorenbos, J., and W. O. Pruitt. 1977. Crop Water Requirements. FAO Irrigation and Drainage Paper No. 24. Rome, Italy: Food and Agriculture Organization of the United Nations.

Dudek, D. J. 1989. Climate change impacts upon agriculture and resources: a case study of California. Appendix C (Agriculture), Vol. 1, Chapter 5 in The Potential Effects of Global Climate Change on the United States. Report EPA-230-05-89-053. Washington, D.C.: U.S. Environmental Protection Agency.

Enoch, H. Z., and B. A. Kimball. 1986. Carbon Dioxide Enrichment of Greenhouse Crops, Vols. I and II. Boca Raton, Florida: CRC Press, Inc.

Geraghty, J. J., D. W. Miller, F. van der Leeder, F. L. Troise, M. Pinther, and R. S. Collins. 1973. Water Atlas of the United States. Port Washington, New York: Water Information Center, Inc.

Gleick, P. H. 1990. Vulnerability of water systems. Pp. 223-240 in P. E. Waggoner, ed., Climate Change and U.S. Water Resources. New York: John Wiley and Sons.

Grotch, S. L. 1988. Regional Intercomparisons of General Circulation Model Prediction and Historical Climate Data--TR041. Report DOE/NBB-0884. Washington, D.C.: U.S. Department of Energy, Carbon Dioxide Research Division.

Guinn, G., and J. R. Mauney. 1980. Analysis of CO₂ exchange assumptions: feedback control. Pp. 1-16 in J. D. Hesketh and J. W. Jones, eds., Predicting Photosynthesis for Ecosystem Models, Vol. II. Boca Raton, Florida: CRC Press, Inc.

Hanks, R. J., H. R. Gardner, and R. L. Florian. 1969. Plant growth-evapotranspiration relationships for several crops in the Great Plains. Agron. J. 61:30-34.

Hansen, J., A. Lacis, D. Rind, G. Russell, P. Stone, I. Fund, R. Ruedy, and J. Lerner. 1984. Climate sensitivity: analysis of feedback mechanisms. Pp. 130-163 in J. E. Hansen and T. Takahashi, eds., Climate Processes and Climate Sensitivity. Geophys. Monog. Ser., Vol. 29. Washington, D.C.: American Geophysical Union.

Hansen, J., I. Fung, A. Lacis, S. Lebedeff, D. Rind, R. Ruedy, G. Russell, and P. Stone. 1988. Global climate change as forecast by the GISS 3-D model. J. Geophys. Res. 93:9341-9364.

Idso, S. B., B. A. Kimball, M. G. Anderson, and J. R. Mauney. 1987. Effects of atmospheric CO₂ enrichment on plant growth: the interactive role of air temperature. Agri. Ecosystems Environ. 20:1-10.

Jensen, M. E. 1966. Empirical methods of estimating or predicting evapotranspiration using radiation. Pp. 57-61, 64 in Proceedings of a Conference on Evapotranspiration and Its Role in Water Resources Management. St. Joseph, Michigan: American Society of Agricultural Engineers.

Jensen, M. E. 1974. Consumptive Use of Water and Irrigation Water Requirements. New York: Technical Committee on Irrigation Water Requirements, Irrigation and Drainage Division, American Society of Civil Engineers.

Jensen, M. E., and H. R. Haise. 1963. Estimating evapotranspiration from solar radiation. Journal of the Irrigation and Drainage Division, American Society of Civil Engineers 89:15-41.

Jones, C. A., and J. R. Kiniry, eds. 1986. CERES-Maize: A Simulation Model of Maize Growth and Development. College Station: Texas A & M University Press.

Jones, J. W., K. J. Boote, S. S. Jagtap, and J. W. Mishoe. 1989. Soybean development. Chapter 5 in R. J. Hanks and J. T. Ritchie, eds., Modeling Plant and Soil Systems. Madison, Wisconsin: American Society of Agronomy.

Jones, P., L. H. Allen, Jr., J. W. Jones, K. J. Boote, and W. J. Campbell. 1984. Soybean canopy growth, photosynthesis, and transpiration responses to whole-season carbon dioxide enrichment. Agron. J. 76:633-637.

Jones, P., L. H. Allen, Jr., and J. W. Jones. 1985a. Responses of soybean canopy photosynthesis and transpiration to whole-day temperature changes in different CO₂ environments. Agron. J. 77:242-249.

Jones, P., L. H. Allen, Jr., J. W. Jones, and R. R. Valle. 1985b. Photosynthesis and transpiration responses of soybean canopies to short-and long-term CO₂ treatments. Agron. J. 77:119-126.

Jones, P., J. W. Jones, and L. H. Allen, Jr. 1985c. Seasonal canopy CO₂ exchange, water use, and yield components in soybean grown under differing CO₂ and water stress conditions. Trans. ASAE 28:2021-2028.

Jouzel, J., C. Lorius, J. R. Petit, C. Genthon, N. I. Barkov, V. M. Kotlyakov, and V. M. Petrov. 1987. Vostok ice core: a continuous isotope temperature record over the last climatic cycle (160,000 years). Nature 329:403-407.

Keeling, C. D., R. B. Bacastow, A. F. Carter, S. C. Piper, T. P. Whorf, M. Heinmann, W. G. Mook, and H. Roeloffzen. 1989. A three dimensional model of atmospheric CO₂ transport based on observed winds: Analysis of data. Pp. 165-234 in D. H. Peterson, ed., Aspects of Climate Variability in the Pacific and the Western Americas. Geophysical Monograph 55. Washington, D.C.: American Geophysical Union.

Kimball, B. A. 1983. Carbon dioxide and agricultural yield: an assemblage and analysis of 430 prior observations. Agron. J. 75:779-788.

Kimball, B. A., and S. B. Idso. 1983. Increasing atmospheric CO₂: effects on crop yield, water use, and climate. Agric. Water Management 7:55-72.

Kimball, B. A., J. R. Mauney, G. Guinn, F. S. Nakayama, P. J. Pinter, Jr., K. L. Clawson, R. J. Reginato, and S. B. Idso. 1983. Response of Vegetation to Carbon Dioxide, Ser. 021: Effects of Increasing Atmospheric CO₂ on the Yield and Water Use of

Crops. Joint program of the U.S. Department of Energy and the U.S. Department of Agriculture, U.S. Water Conservation Lab, and U.S. Western Cotton Research Lab. Phoenix, Arizona: U.S. Department of Agriculture, Agricultural Research Service.

King, G. A., R. L. DeVelice, R. P. Neilson, and R. C. Worrest. 1989. Pp. 251-285 (Chapter 4) in The Potential Effects of Global Climate Change on the United States. Report EPA-230-05-89-050. Washington, D.C.: U.S. Environmental Protection Agency.

Knisel, W. G., P. Yates, J. M. Sheridan, T. K. Woody, L. H. Allen, Jr., and L. E. Asmussen. 1985. Hydrology and Hydrogeology of Upper Taylor Creek Watershed, Okeechobee County, Florida: Data and Analysis. Report ARS-25. Washington, D.C.: U.S. Department of Agriculture, Agricultural Research Service.

Landers, H. 1974. The climate of South Carolina. Pp. 353-369 in Climate of the States, Vol. 1. Port Washington, New York: Water Information Center, Inc.

Lorius, C., J. Jouzel, D. Raynaud, J. Hansen, and H. Le Treut. 1990. The ice core record: climate sensitivity and future greenhouse warming. Nature 347:139-145.

Manabe, S., and R. T. Wetherald. 1986. Reduction in summer soil wetness induced by an increase in atmospheric carbon dioxide. Science 232:626-628.

Manabe, S., and R. T. Wetherald. 1987. Large-scale changes of soil wetness induced by an increase in atmospheric carbon dioxide. J. Atmos. Sci. 44:1211-1235.

Mearns, L. O. 1989. Climate variability. Pp. 29-55 (Chapter 3) in The Potential Effects of Global Climate Change on the United States. Report EPA-230-05-89-050. Washington, D.C.: U.S. Environmental Protection Agency.

Miller, Barbara A., and W. G. Brock. 1988. Sensitivity of the Tennessee Valley Authority Reservoir System to Global Climate Change. Report No. WR28-1-680-101. Norris, Tennessee: Tennessee Valley Authority.

Miller, Barbara., and W. G. Brock. 1989. Global climate change: implications for the Tennessee Valley Authority reservoir system. Pp. 493-500 in J. C. Topping, Jr., ed., Coping with Climate Change: Proceedings of the Second North American Conference on Preparing for Climate Change. Washington, D.C.: The Climate Institute.

Mitchell, J. F. B. 1989. The "greenhouse" effect and climate change. Reviews of Geophysics 27:115-139.

Morison, J. I. L. 1987. Intercellular CO₂ concentration and stomatal response to CO₂. Pp. 229-252 in E. Zeiger, G. D.

Farquhar, and I. R. Cowan, eds., Stomatal Function. Stanford, California: Stanford University.

Morison, J. I. L., and R. A. Gifford. 1984. Plant growth and water use with limited water supply in high CO₂ concentrations: I. leaf area, water use, and transpiration. Australian J. Plant Physiol. 11:361-374.

Pearcy, R. W., and O. Björkman. 1983. Physiological effects. Pp. 65-105 in E. R. Lemon, ed., Carbon Dioxide and Plants: The Response of Plants to Rising Levels of Atmospheric Carbon Dioxide. American Association for the Advancement of Science Selected Symposium 84. Boulder, Colorado: Westview Press.

Peart, R. M., J. W. Jones, R. B. Curry, K. J. Boote, and L. H. Allen, Jr. 1989. Impact of climate change on crop yield in the Southeastern USA: a simulation study. Appendix C (Agriculture), Vol. 1, Chapter 2 in J. B. Smith and D. A. Tirpak, eds., The Potential Effects of Global Climate Change on the United States. Report EPA-230-05-89-053. Washington, D.C.: U.S. Environmental Protection Agency.

Penning de Vries, F. W. T., D. M. Jansen, H. F. M. ten Berge, and A. Bakema. 1989. Simulation of Ecophysiological Processes of Growth in Several Annual Crops. Wageningen, Netherlands: Centre for Agricultural Publication and Documentation (Pudoc).

Revelle, R., and P. W. Waggoner. 1983. Effects of a carbon dioxide induced climatic change on water supplies in the Western United States. Pp. 419-432 in Changing Climate: Report of the Carbon Dioxide Assessment Committee. Washington, D.C.: National Academy Press.

Ritchie, J. T., B. D. Baer, and T. Y. Chou. 1989. Effect of global climate change on agriculture: Great Lakes region. Appendix C (Agriculture), Vol. 1, Chapter 1 in J. B. Smith and D. A. Tirpak, eds., The Potential Effects of Global Climate Change on the United States. Report EPA-230-05-89-053. Washington, D.C.: U.S. Environmental Protection Agency.

Rosenberg, N. J., B. A. Kimball, P. Martin, and C. F. Cooper. 1990. From climate and CO₂ enrichment to evapotranspiration. Pp. 151-175 in P. E. Waggoner, ed., Climate Changes and U.S. Water Resources. New York: John Wiley and Sons.

Rosenzweig, C. 1989. Potential effects of climate change on agricultural production in the Great Plains: a simulation study. Appendix C (Agriculture), Vol. 1, Chapter 3 in J. B. Smith and

D. A. Tirpak, eds., The Potential Effects of Global Climate Change on the United States. Report EPA-230-05-89-053. Washington, D.C.: U.S. Environmental Protection Agency.

Schlesinger, M. E. 1984. Climate model simulation of CO₂-induced climate change. Pp. 141-235 in B. Saltzman, ed., Advances in Geophysics, Vol. 26. New York: Academic Press.

Shawcraft, R. W., E. R. Lemon, L. H. Allen, Jr., D. W. Stewart, and S. E. Jensen. 1974. The soil-plant-atmosphere model and some of its predictions. Agric. Meteorol. 14:287-307.

Smith, J. B., and D. A. Tirpak, eds. 1989. The potential effects of global climate change on the United States. Report EPA-230-05-89-050. Washington, D.C.: U.S. Environmental Protection Agency.

Stanhill, G. 1960. The relationship between climate and the transpiration and growth of pastures. Pp. 293-296 in P. J. Boyle, and L. W. Raymond, eds., Proceedings, Eighth International Grasslands Congress, University of Reading, Berkshire, England. Oxford, England: Alden Press.

Stephens, J. C. 1965. Estimating evaporation from insolation. J. Hydr. Div., Am. Soc. Civ. Engr. 91(HY5):171-182.

Stephens, J. C., and E. H. Stewart. 1963. A Comparison of Procedures for Computing Evaporation and Evapotranspiration. Pub. No. 62, Intern. Assoc. of Sci. Hydrol. Berkeley, California: Trans. Intern. Union Geodesy and Geophysics.

Tanner, C. B., and T. R. Sinclair. 1983. Efficient water use in crop production: research or research? Pp. 1-27 in H. M. Taylor, W. R. Jordan, and T. R. Sinclair, eds., Limitations to Efficient Water Use in Crop Production. Madison, Wisconsin: American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America.

Tolbert, N. E., and I. Zelitch. 1983. Carbon metabolism. Pp. 21-64 in E. R. Lemon, ed., CO₂ and Plants: The Response of Plants to Rising Levels of Atmospheric Carbon Dioxide. American Association for the Advancement of Science Selected Symposium 84. Boulder, Colorado: Westview Press.

Waggoner, P. E. 1990. Climate Change and U.S. Water Resources. New York: John Wiley & Sons.

Warrick, R. A., R. M. Gifford, and M. L. Parry. 1986. CO₂, climate change, and agriculture. Pp. 393-473 in B. Bolin, B. R. Doos, J. Jager, and R. A. Warrick, eds., The Greenhouse Effect, Climate Change, and Ecosystems (SCOPE 29). New York: John Wiley and Sons.

Washington, W. M., and G. A. Meehl. 1983. General circulation model experiments on the climatic effects due to doubling and

quadrupling of carbon dioxide concentrations. J. Geophys. Res. 88:6600-6610.

Washington, W. M., and G. A. Meehl. 1984. Seasonal cycle experiment on the climate sensitivity due to a doubling of CO₂ with an atmospheric general circulation model coupled to a simple mixed layer ocean model. J. Geophys. Res. 89:9475-9503.

Washington, W. M., and G. A. Meehl. 1986. General circulation model CO₂ sensitivity experiments: snow-sea ice albedo parameterizations and globally averaged surface air temperature. Clim. Change 8:231-241.

Wilkerson, G. G., J. W. Jones, K. J. Boote, K. T. Ingram, and J. W. Mishoe. 1983. Modeling soybean growth for crop management. Trans. ASAE 26:53-73.

Wilson, C. A., and J. F. B. Mitchell. 1987. A doubled CO₂ climate sensitivity experiment with a GCM including a simple ocean. J. Geophys. Res. 92:13315-13343.

Zazueta, F. S., and A. G. Smajstrla. 1989. Water Management Utilities. Pub. 89-1, Fla. Coop. Ext. Serv., Institute of Food and Agricultural Sciences. Gainesville: University of Florida.

What Decreases Carbon Dioxide Levels in the Atmosphere

Source: https://www.nap.edu/read/1911/chapter/8

What Decreases Carbon Dioxide Levels in the Atmosphere

7
Effects of Increasing Carbon Dioxide Levels and Climate Change on Plant Growth, Evapotranspiration, and Water Resources

BACKGROUND: PLANT RESPONSES TO ENVIRONMENTAL FACTORS

DIRECT EFFECTS OF CARBON DIOXIDE ON PHOTOSYNTHESIS, TRANSPIRATION, AND GROWTH OF PLANTS

Atmospheric Carbon Dioxide

Plant Photosynthetic Mechanisms

Plant Growth Responses to Carbon Dioxide

Transpiration Responses to Carbon Dioxide

Streamflow Responses to Carbon Dioxide

Plant Water-Use Efficiency

CLIMATE CHANGE EFFECTS ON PHOTOSYNTHESIS, GROWTH, AND TRANSPIRATION

INTERACTIVE EFFECTS OF CARBON DIOXIDE AND CLIMATE CHANGE

Photosynthetic and Productivity Interactions

Evapotranspiration

ESTIMATING YIELD AND WATER REQUIREMENTS OF CROPS UNDER TWO CURRENT CLIMATE CHANGE SCENARIOS

General Circulation Models

Modeling Crop Responses to Carbon Dioxide and Climate Changes

IMPLICATIONS OF CHANGING EVAPOTRANSPIRATION AND PRECIPITATION FOR WATER RESOURCES

Adaptations and Evapotranspiration Requirements

Water Availability

SUMMARY

ACKNOWLEDGMENTS

REFERENCES

0 Response to "What Decreases Carbon Dioxide Levels in the Atmosphere"

Post a Comment

Iklan Atas Artikel

Iklan Tengah Artikel 1

Iklan Tengah Artikel 2

Iklan Bawah Artikel

What Decreases Carbon Dioxide Levels in the Atmosphere

7 Effects of Increasing Carbon Dioxide Levels and Climate Change on Plant Growth, Evapotranspiration, and Water Resources

BACKGROUND: PLANT RESPONSES TO ENVIRONMENTAL FACTORS

DIRECT EFFECTS OF CARBON DIOXIDE ON PHOTOSYNTHESIS, TRANSPIRATION, AND GROWTH OF PLANTS

Atmospheric Carbon Dioxide

Plant Photosynthetic Mechanisms

Plant Growth Responses to Carbon Dioxide

Transpiration Responses to Carbon Dioxide

Streamflow Responses to Carbon Dioxide

Plant Water-Use Efficiency

CLIMATE CHANGE EFFECTS ON PHOTOSYNTHESIS, GROWTH, AND TRANSPIRATION

INTERACTIVE EFFECTS OF CARBON DIOXIDE AND CLIMATE CHANGE

Photosynthetic and Productivity Interactions

Evapotranspiration

ESTIMATING YIELD AND WATER REQUIREMENTS OF CROPS UNDER TWO CURRENT CLIMATE CHANGE SCENARIOS

General Circulation Models

Modeling Crop Responses to Carbon Dioxide and Climate Changes

IMPLICATIONS OF CHANGING EVAPOTRANSPIRATION AND PRECIPITATION FOR WATER RESOURCES

Adaptations and Evapotranspiration Requirements

Water Availability

SUMMARY

ACKNOWLEDGMENTS

REFERENCES

0 Response to "What Decreases Carbon Dioxide Levels in the Atmosphere"

Post a Comment

Iklan Atas Artikel

Iklan Tengah Artikel 1

Iklan Tengah Artikel 2

Iklan Bawah Artikel

7
Effects of Increasing Carbon Dioxide Levels and Climate Change on Plant Growth, Evapotranspiration, and Water Resources