4. The CO₂ fertilization effect: higher carbohydrate production and retention as biomass and seed yield

L. HARTWELL ALLEN, JR.
US Department of Agriculture, Agricultural Research Service, University of Florida, Gainesville, Florida, USA

JEFF T. BAKER AND KEN J. BOOTE
Agronomy Department, University of Florida, Gainesville, Florida, USA

Overview of CO₂ effects on plant growth processes
Specific responses of crops to elevated CO₂
Summary of comprehensive reviews
Crop modelling: Predictions for the future
Summary and conclusions
Acknowledgements
References

The rise in atmospheric carbon dioxide (CO₂) concentration from about 280 m mol/mol before the industrial revolution to about 360 m mol/mol currently is well documented (e.g., Baker and Enoch, 1983; Keeling et al., 1995). The consensus of many studies of the effects of elevated CO₂ on plants is that the CO₂ fertilization effect is real (see Kimball, 1983; Acock and Allen, 1985; Cure and Acock, 1986; Allen, 1990; Rozema et al., 1993; Allen, 1994; Allen and Amthor, 1995). However, the CO₂ fertilization effect may not be manifested under conditions where some other growth factor is severely limiting, such as low temperature (Long, 1991). Also, plants grown in some conditions, where limitations of rooting volume (Arp, 1991), light, or other factors restrict growth, have not shown a sustained response to elevated CO₂ (Kramer, 1981).

The main objectives of this chapter are to assess the direct effects of rising atmospheric CO₂ and indirect effects of potential climate changes on crop growth and yield. The approach will be to (1) provide a general overview of CO₂ effects on plant growth processes; (2) analyse some specific experimental data on crop plant responses to elevated CO₂ and climate change factors; (3) summarize recent reviews of plant responses to elevated CO₂; and (4) discuss some crop modelling assessments of rising CO₂ and climate change factors on agricultural productivity based on predictions of global climate change models. Also, some adaptations for improving crop productivity in a higher CO₂ world will be suggested.

Overview of CO₂ effects on plant growth processes

Most of the following discussion of CO₂ effects on plants applies to species with the C₃ photosynthetic pathway and not necessarily to species with the C₄ pathway. Other aerial, non-biotic environmental factors that affect plant growth and development are light and temperature. Plant photosynthetic rates generally increase linearly with light across relatively low ranges of light intensity, and then the rates decelerate until they reach an asymptotic maximum. Because of crowding and shading of many leaves, most crop canopies do not reach light saturation at full sunlight; that is, they would be able to respond to light levels well beyond full solar irradiance. Likewise, crop photosynthetic rates respond to increasing levels of CO₂ but then level off at higher concentrations (around 700 m mol/mol or greater, depending upon species and other factors). However, leaf photosynthesis usually increases with temperature up to some maximum value, and then declines. Furthermore, temperature affects not only photosynthesis, but also respiration, growth, development phases and reproductive processes.

Elevated CO₂ may have some effects on crop phenology, although stages of development are governed primarily by temperature, time and photoperiod. If dates of planting were to be changed because of the greenhouse effect, then phenological timing of plants could be affected. For example, higher temperatures could decrease yields by decreasing the duration of the grain-filling period or changes in photoperiod could shorten or lengthen the vegetative stage.

The CO₂ fertilization effect begins with enhanced photosynthetic CO₂ fixation. Non-structural carbohydrates tend to accumulate in leaves and other plant organs as starch, soluble carbohydrates or polyfructosans, depending on species. In some cases, there may be feedback inhibition of photosynthesis associated with accumulation of non-structural carbohydrates. Increased carbohydrate accumulation, especially in leaves, may be evidence that crop plants grown under CO₂ enrichment may not be fully adapted to take complete advantage of elevated CO₂. This may be because the CO₂-enriched plants do not have an adequate sink (inadequate growth capacity), or lack capacity to load phloem and translocate soluble carbohydrates. Improvement of photoassimilate utilization should be one goal of designing cultivars for the future (Hall and Allen, 1993).

In the process of growth, photoassimilates are allocated to the vegetative shoots, root system or reproductive organs. In some cases, more photoassimilate of CO₂-enriched plants is partitioned to the root system than to the shoots. Above ground, more photoassimilate usually goes into stems and supporting structures than into leaves. This phenomenon may not be an inherent response to elevated CO₂, but may be a by-product of the larger size of plants often found in CO₂-enriched atmospheres, especially by species that produce branch stems along the aerial mainstems (Allen et al., 1991).

Reproductive biomass growth as well as vegetative biomass growth are usually increased by elevated CO₂. However, the harvest index, or the ratio of seed yield to above-ground biomass yield, is typically lower under elevated CO₂ conditions (Allen, 1991; Baker et al., 1989), which may also be evidence of the lack of capacity to utilize completely the more abundant photoassimilate.

In many cases, both the amount and the carboxylation activity of ribulose 1,5-bisphosphate carboxylase-oxygenase enzyme (rubisco) is decreased in leaves of plants grown under elevated CO₂ This acclimation phenomenon may produce 'downregulation of photosynthesis'; however, this is not universally the case. For example, there is little evidence of this downregulation response in soybean (Glycine max L. Merr.), a C₃ legume. In fact, photosynthetic capacity per unit leaf area of soybean is increased under CO₂ enrichment. Leaves often develop an additional layer of mesophyll cells. Also, more structural carbohydrate may be produced in leaves, as well as stems, of CO₂-enriched plants.

We can obtain clues about reasons that some plants downregulate and others do not in response to elevated CO₂ by focusing on the capabilities of soybean. This plant has (a) symbiotic N2 fixation; (b) the capacity to form additional layers of palisade cells in the leaf tissue; (c) the capacity to shunt much of the photoassimilate into relatively inert starch rather than soluble sugars during photosynthesis; (d) a relatively strong leaf and stem sink during vegetative development; and (e) a strong seed-fill sink during reproductive development. Plants which lack these capacities, either inherently or because of growth in limiting environments, are more likely to demonstrate some degree of downregulation of photosynthesis (Allen, 1994).

The carbon:nitrogen ratio of leaves of plants is usually increased under CO₂ enrichment. Plants may acclimate to elevated CO₂ by requiring less rubisco and photo-synthetic apparatus, which would lead to lower nitrogen contents. The overall change in C:N ratios is governed both by increases in structural and non-structural carbohydrates, and by decreases in protein content. However, seed nitrogen content is little affected (Allen et al., 1988).

Specific respiration rates may be reduced by both short-term exposure to elevated CO₂ and long-term growth at elevated CO₂ (Amthor, 1995). However, the long-term effect may be similar when respiration rates are reported on a per unit nitrogen basis.

In climate change scenarios, temperatures are predicted to increase following the rise of CO₂ and other greenhouse-effect gases. Carbon dioxide x temperature interactions have been observed for vegetative growth (i.e., the CO₂ fertilization effect is greater at warmer temperatures than at cooler temperatures). Temperature increases in a higher CO₂ world could increase overall biomass productivity for vegetative crops (pastures and forages) both by extending the length of the growing season in temperate regions, and by the interaction of CO₂ x temperature in stimulation of vegetative growth. However, CO₂ x temperature interactions appear to be very small or negligible for reproductive processes (seed set and seed yield) although there may be more initial flowers formed by greater amounts of branching or tillering that is stimulated by CO₂ enrichment (Baker and Allen, 1993a).

Precipitation changes may occur along with other climatic change effects. In general, predictions from crop models show that increased CO₂ should increase productivity of C₃ plants, but the associated predictions of temperature rise will be detrimental. Not surprisingly, changes in precipitation patterns (decreases of rainfall during growth period) could be more detrimental for crop production than changes in temperature.

Under elevated CO₂ stomatal conductance in most species will decrease which may result in less transpiration per unit leaf area. However, leaf area index of some crops may also increase. The typical 40% reduction in stomatal conductance induced by a doubling of CO₂ has generally resulted in only a 10% (or less) reduction in crop canopy water use in chamber or field experimental conditions. Actual changes in crop evapotranspiration will be governed by the crop energy balance, as mitigated by stomatal conductance, leaf area index, crop structure and any changing meteorological factors.

Water-use efficiency (WUE) (ratio of CO₂ uptake to evapotranspiration) will increase under higher CO₂ conditions. This increase is caused more by increased photosynthesis than it is by a reduction of water loss through partially closed stomata. Thus, more biomass can be produced per unit of water used, although a crop would still require almost as much water from sowing to final harvest. If temperatures rise, however, the increased WUE caused by the CO₂ fertilization effect could be diminished or negated, unless planting dates can be changed to more favourable seasons.

Several assessments of impacts of climate change on crop productivity have been published. Progress has been made on integrating the impacts on individual countries and on economic and social interactions. For the most part, these assessments project more favourable climates for agriculture in northern latitudes and less favourable climates in the tropical and subtropical zones. However, the crop modelling predictions are dependent on the scenarios of outputs of General Circulation Models (GCMs) applied to the greenhouse effect. Thus, the dependency chain of assessments follows: Climate Change Scenarios ® Crop Model Prediction and Agricultural Production Systems (with and without available mitigation and adaptation response strategies) ® National Scenarios of Economics and Well-being of Farmers, Agricultural Commerce, and Consumers ® Country-by-Country and Global Interaction Scenarios of World Trade (food and all other commodities). Population Dynamics and Economic Well-being, and Impacts on Social Systems.

As the world continues to consume fossil fuels, CO₂ concentrations will continue to rise. Other greenhouse-effect gases, such as methane, nitrous oxides, chlorofluorocarbons and chlorofluorocarbon substitutes, and perhaps tropospheric ozone, will likely rise also. The CO₂ fertilization effect on plants will increase and climate changes may occur because of the combined increase of all greenhouse-effect gases. Global agriculture could adapt to gradual regional climate changes, but sudden changes would be more serious. Adaptation and/or mitigation actions could include the following:

1. Selection of plants that can better utilize carbohydrates which are produced when plants are grown at elevated CO₂.

2. Selection of plants that produce less structural matter and more reproductive capacity under CO₂ enrichment. (This applies for seed crop plants, not necessarily vegetative biomass plants.)

3. Search for germplasms that are adapted to higher day and night temperatures, and incorporate those traits into desirable crop production cultivars to improve flowering and seed set.

4. Change planting dates and other crop management procedures to optimize yields under new climatic conditions, and select for cultivars that are adapted to these changed agricultural practices.

5. Shift to species that have more stable production under high temperatures or drought.

6. Determine whether more favourable N:C ratios can be attained in forage cultivars adapted to elevated CO₂.

7. Where needed, and where possible, develop irrigation systems for crops.

Specific responses of crops to elevated CO₂

This section will focus on responses of two C₃ crops to elevated CO₂: soybean, a symbiotic nitrogen-fixing legume representative of the pulses; and rice (Oryza sativa L.), a globally important cereal food crop. Findings from seven crop cycles of both soybean and rice grown in sunlit chambers at Gainesville, Florida, USA, will be emphasized (Baker and Allen, 1993a,b). The chambers feature both large rooting volumes of real soil and light from the sun. Some studies included responses to sub-ambient as well as superambient CO₂ concentrations (e.g., Allen et al., 1991; Baker and Allen, 1993a,b, 1994). (The rice experiments have been designated chronologically as RICE I, II, III, IV, V, VI and VII.)

SOYBEAN

Photosynthetic rates

Soybean photosynthetic rates of both leaves and the whole canopy have always been increased by elevated CO₂ Jones et al. (1984) reported midday maximum canopy photosynthetic rates of 60 and 90 m mol CO₂/m²/s at 70 days after planting (DAP) for soybean grown at CO₂ concentrations of 330 and 800 m mol/mol, respectively. The leaf area index (LAI) was 6.9 and 9.0, respectively. Midday maximum canopy photo-synthetic rates in other studies were 40 and 75 m mol/m²/s for soybean grown at 320 and 640 m mol/mol CO₂ (Jones et al., 1985c) and 40 and 80 m mol/m²/s for soybean grown at 330 and 800 m mol/m²/s (Jones et al., 1985a,b).

Valle et al. (1985) found that midday maximum photosynthetic CO₂ uptake rates of soybean leaves ranged from 30 to 50 m mol/m²/s and 15 to 25 m mol/m²/s on plants grown at 660 and 330 m mol/mol CO₂ respectively. Allen et al. (1990) reported that, at all light levels, leaf photosynthetic rates increased linearly with CO₂ concentration across the range of 330 to 800 m mol/mol.

Valle et al. (1985) used a Michaelis-Menten type of rectangular hyperbola to summarize photosynthetic responses of soybean leaves vs. CO₂ concentration. The plants had been grown at 330 and 660 m mol/mol of CO₂ and then exposed to a wide range of CO₂ for a short period.

Y=(Y_max x [C])/([C]+K_m)+Y_i (4.1)

where Y is photosynthetic rate in m mol CO₂/m²/s; [C₃] is CO₂ concentration in m mol/mol; Y_i is the y-axis intercept at zero [C₃], the apparent respiration rate, in m mol CO₂/m²/s; Y_max is the response limit of (Y - Y_i) at very high [C], the asymptotic photosynthetic rate, in m mol CO₂/m²/s; K_m is the value of [C] where (Y-Y_i)=Y_max/2, the apparent Michaelis-Menten constant, in (m mol/mol; and G _c is the calculated [C] intercept at zero Y, the CO₂ compensation point, m mol/mol (not shown in this equation). The average parameters for responses at 330 and 660 m mol/mol are given in Table 4.1. There was no obvious downregulation of soybean leaf photosynthesis in response to elevated CO₂; in fact, photosynthetic capacity was increased. Leaf quantum yield increased from 0.05 to 0.09 in the soybean leaves exposed to CO₂ of 330 and 660 m mol/mol, respectively (Valle et al., 1985).

Campbell et al. (1988) showed that soybean leaf photosynthetic rates were higher for plants grown at 660 than at 330 m mol/mol CO₂ when measured at common intercellular CO₂ concentrations. Furthermore, Campbell et al. (1988) measured rubisco activity and amount in leaves of soybean grown in CO₂ concentrations of 160, 220, 280, 330, 660 and 990 m mol/mol. They found that rubisco activity was almost constant at 1.0 m mol CO₂/min/mg soluble protein across this CO₂ treatment range. Leaf soluble protein was nearly constant at about 2.4 g/m² with 55% being rubisco protein. Specific leaf weight increased across the 160 to 990 m mol/mol CO₂ concentration range, so that the rubisco activity on a leaf dry weight basis decreased.

Campbell et al. (1990) showed that photosynthetic capacity of soybean canopies grown at 330 and 660 m mol/mol CO₂ were similar for short exposures at concentrations below 500 m mol/mol, but above this concentration, canopies grown at 660 m mol/mol had a slightly higher photosynthetic capacity. Thus, soybean did not lose photo-synthetic capacity as did some other plant species (see Allen, 1994). In other studies, however, both photosynthetic rate and rubisco activity of soybean declined during long-term CO₂ enrichment (Thorne and Koller, 1974; Delucia et al., 1985).

Respiration

Whole-canopy pre-dawn respiration rates at 40 to 60 DAP were 2-3 and 4-5 m mol/m²/s for soybean grown at CO₂ levels of 320 and 640 m mol/m², respectively, and a constant day/night temperature of 25°C (Jones et al., 1985c). After cross-switching two of the treatment chambers at DAP 52, the subsequent canopy photosynthetic rates and respiration rates quickly adjusted to their new CO₂ exposure conditions.

Table 4.1. Average asymptotic maximum photosynthetic rate (Y_max) with respect to y-intercept parameter (Y_i), apparent Michaelis-Menten constant for CO₂ (K_m, and CO₂ compensation point (G _c.) for leaves grown at two CO₂ treatments and subjected to different short-term CO₂ levels. Condensed from Valle et al. (1985)

Growth CO2 treatment	Ymax m mol/m2/s	Km m mol/mol	Yi m mol/m2/s	G c m mol/mol
330	51.8	359	-7.8	63
660	126.6	1 133	-4,6	42

Means of Y_max and Y_i were significantly different, p = 0.05, by a t-test.

Thus, pre-dawn respiration rates were closely connected to the previous CO₂ fixation rates.

Partitioning

Growth of plants under elevated CO₂ results in changes in partitioning of photoassimilates to various plant organs over time (Table 4.2). In soybean, elevated CO₂ generally promoted greater carbon (dry matter) partitioning to the supporting structure (stems, petioles and roots) than to the leaf laminae during vegetative stages of growth (Allen et al., 1991). During reproductive stages, there tended to be lower relative partitioning to reproductive growth (pods) by plants under elevated CO₂.

Table 4.2. Soybean plant components as a percentage of total dry matter grown at subambient and superambient concentrations of CO₂ in 1984. Condensed from Allen et al. (1991).

Component	CO2 concentration, m mol/mol
	160	220	280	330	660	990
	13 DAP 1 (V2 Stage)
Root, %	11.3	9.8	9.3	12.4	9.8	10.5
Cotyledon, %	19.5	15.6	13.9	13.8	11.2	9.4
Stem, %	22.8	25.0	25.8	25.5	26.3	26.7
Leaf, %	46.1	49.5	50.7	48.2	52.5	53.2
	34 DAP (V8 Stage)
Root, %	6.4	7.1	6.9	7.6	8.9	7.9
Stem, %	23.4	24.4	27.6	25.9	28.2	29.3
Petiole, %	12.2	13.1	14.6	14.4	15.7	15.7
Leaf, %	57.6	55.3	50.8	52.2	47.2	46.9
	66 DAP (R5 Stage)
Stem, %	16.5	19.6	25.0	21.7	25.2	26.8
Petiole, %	9.9	11.3	12.9	12.7	14.1	13.6
Leaf, %	35.3	35.3	35.4	32.2	31.5	29.8
Pod, %	38.2	33.6	26.5	33.3	28.9	29.6
	94 DAP (R7 Stage)
Stem, %	10.2	12.8	14.9	14.3	16.3	19.5
Petiole, %	5.5	6.2	6.0	6.7	7.0	6.5
Leaf, %	17.4	16.3	14.9	14.4	11.9	10.9
Pod, %	66.7	64.7	64.1	64.6	64.7	64.0

¹ Days after planting.

Growth rates

During the linear phase of vegetative growth after full ground cover is reached, the growth rates of plants exposed to a range of CO₂ concentrations varied from 5.0 to 20.7 g/m²/d for exposures from 160 to 990 m mol/mol (Allen et al., 1991). The total final dry weight ranged from 12.88 to 39.12 g/plant, and final seed weight ranged from 5.77 to 17.85 g/plant for CO₂ treatments ranging from 160 to 660 m mol/mol.

Carbohydrates

Soybean accumulates non-structural carbohydrates, particularly starch, under CO₂ enrichment. Allen et al. (1988) grew soybean under CO₂ treatments of 330, 450, 600 and 800 m mol/mol. Sucrose, reducing sugars, and total soluble sugars of leaves remained somewhat constant throughout the day, but starch increased steadily at a rate of about 6 g/kg dry matter/hour. Average total soluble sugars increased from 24 to 36 g/kg dry weight and starch increased from 85 to 204 g/kg dry weight across the range of 330 to 800 m mol/mol. Elevated non-structural carbohydrates in CO₂-enriched soybean plants were confirmed by Baker et al. (1989) and Allen et al. (1995). The concentrations also varied across the life cycle of the plants.

Nitrogen

Nitrogen content decreased from 50 to 37 g/kg dry weight over the range of 330 to 800 m mol/mol. When the nitrogen content was adjusted to remove the effect of total non-structural carbohydrate, the relative changes were smaller, 55 to 48 g N/kg dry matter, across the 330 to 800 m mol/mol range (Allen et al., 1988).

Yield

Soybean seed yield was always increased by elevated CO₂ Allen et al. (1987) summarized the photosynthetic, biomass and seed yield responses of several experiments with the equation (4.1) rectangular hyperbola model using data normalized to responses obtained at 330 m mol/mol. The values of K_m , Y_max and Y_i parameters for relative photosynthetic rates were 279 m mol/mol, 3.08 and -0.68, respectively, for relative biomass yield were 182 m mol/mol, 3.02 and -0.91, respectively, and for relative seed yield were 141 m mol/mol, 2.55 and -0.76, respectively. This model was used to project yields across several ranges of atmospheric CO₂ concentration increases (Table 4.3). For a doubling of CO₂ this model predicted a 32.2% increase in soybean grain yield and a 42.7% increase in biomass. The ratio of these two numbers, 1.322/1.427 = 0.926, gives the fraction of the harvest index expected under doubled CO₂ in comparison with ambient CO₂.

Table 4.3. Percentage increases of soybean midday photosynthetic rates, biomass yield, and seed yield predicted across selected carbon dioxide concentration [CO₂] ranges associated with relevant benchmark points in time. Adapted from Allen et al. (1987)

Period of time (years)	[CO2]-Midday		Biomass photosynthesis	Seed yield	Biomass yield
	Initial	Final
	(Nmd/mol)		(% increase over initial [CO2])
IA-1700 1	200	270	38	33	24
1700-1973	270	330	19	16	12
1973-2073? 2	330	660	50	41	31

¹ IA, the Ice Age about 13 000 to 30 000 years before present. The atmospheric CO₂ concentrations that prevailed during the last Ice Age, and from the end of the glacial melt until pre-pioneer/pre-industrial revolution times, were 200 and 270 m mol/mol, respectively.

² The first world energy 'crisis' occurred in 1973 when the CO₂ concentration was 330 m mol/mol. This CO₂ concentration is used as the basis for many CO₂ doubling studies. The CO₂ concentration is expected to double sometime within the 21st century.

Carbon dioxide and temperature

The CO₂ fertilization effect appears to be enhanced under elevated temperatures, at least up to a point. Idso et al. (1987) and Kimball et al. (1993) showed that the growth modification factor (or biomass growth modification ratio) due to a 300 m mol/mol enrichment was 0.08 per °C (average daily temperature) across the range of 12 to 34°C. However, for soybean, Allen (1991) calculated a biomass growth modification ratio response to temperature of -0.031 per °C for seed biomass yield and -0.026 per °C for total biomass accumulation. Elevated temperatures tended to shorten the grain-filling period of this crop.

Soybean seed yield tended to decrease slightly with temperature over the day/night range of 26/19 to 36/29°C (Table 4.4). The number of seed per plant increased slightly with increase of both CO₂ and temperature. Mass per seed decreased sharply with increasing temperature. Although CO₂ enrichment resulted in increased seed yield and above-ground biomass, harvest index was decreased with both CO₂ and temperature (Baker et al., 1989). The data of Table 4.4 show no tendency for the growth modification factor to increase with temperature for either seed yield or biomass accumulation.

Subsequent experiments with soybean substantiate these data although it was found that seed yields dropped sharply for day/night temperatures of 40/30°C and above, and biomass yields were maintained up to 44/34°C before rapidly failing (Deyun Pan et al., unpublished).

The sensitivity of the growth modification factor data of Idso et al. (1987) and Kimball et al. (1993) may have been affected by other factors at Phoenix, Arizona. Apparently, the growth vs. temperature data were obtained throughout the year, and the findings may have been impacted by factors such as total solar radiation, photoperiod, stages of development, or other conditions caused by the changing seasons. Of course, these other environmental factors are all part of the complex of plant responses to climatic conditions. However, a better test of temperature effects alone would be CO₂ enrichment throughout the season or life cycle of plants under natural conditions, but with consistent temperature differences (cooler and warmer).

Table 4.4. Seed yield, components of yield, total above-ground biomass and harvest index of soybean grown at two CO₂ concentrations and three temperatures in 1987 (adapted from Baker et al., 1989)

CO2 conc. (m mol/mol)	Day/night temperature (°C)	Grain yield (g/plant)	Seed/plant (no./plant)	Seed mass (mg/seed)	Above-ground biomass (g/plant)	Harvest index
330	26/19	9.0	44.7	202	17.1	0.53
330	31/24	10.1	52.1	195	19.8	0.51
330	36/29	10.1	58.9	172	22.2	0.45
660	26/19	13.1	58.8	223	26.6	0.49
660	31/24	12.5	63.2	198	27.6	0.45
660	36/29	11.6	70.1	165	26.5	0.44

F-values
CO2 conc.	12.3**	11.4**	2.5*	NA	NA
Temperature	0.0 NS	8.4**	106.2**	NA	NA
CO2 x Temperature	2.0 NS	0.1 NS	11.2**	NA	NA

*, ** Significant at the 0.05 and 0.01 probability levels, respectively.
NS = not significant; NA = not available.

Evapotranspiration and water-use efficiency

The direct effect of increasing temperatures across the range of 28 to 35°C appears to increase transpiration rate about 4 to 5% per °C, based on both experimental and modelling studies (Allen, 1991). This is in close agreement with the rise in saturation vapour pressure of about 6% per °C. Allen et al. (1985) showed that the increase in WUE of soybean contributed by increased photosynthesis under elevated CO₂ was much greater than the increase in WUE contributed by decreased transpiration. The fractional increase in WUE attributable to increased photosynthesis and decreased transpiration were about 0.8 and 0.2, respectively (based on data of Jones et al., 1985b).

RICE

Leaf and canopy photosynthesis

Leaf photosynthetic rates, in high light, of cv. IR72 rice, determined in 1992 (RICE VI experiment) for plants grown at three day/night temperature regimes, 32/23, 35/26 and 38/29°C (but measured at ambient temperature), averaged 18.8 and 30.4 m mol CO₂/m²/s for 330 and 660 m mol/mol CO₂ treatments, respectively (Allen et al., 1995).

This increase caused by elevated CO₂ is about 60%. Leaf A/C_i curves obtained from both CO₂ treatments at the growth temperature of 35/26°C are shown in Figure 4.1. There were essentially no differences in A between the two CO₂ treatments at each specific C. level. Similar pairs of curves (not shown) were obtained for the other temperature treatments. Each pair of leaf A/C_i curves were similar within each temperature regime. Thus, the leaf A/C_i curves gave no indication of a change in photo-synthetic capacity. Extrapolations of the response curves to zero values of A gave an intercept (leaf CO₂ compensation point) of about 60 m mol/mol CO₂.

Following canopy closure, the rice canopy net photosynthetic rate (Pn) vs. photo-synthetic photon flux density (PPFD) responses were linear and did not approach light saturation in any of the experiments, probably because of the erect leaf orientation of cv. IR30 and high plant populations. Canopy Pn vs. PPFD at 60 DAP for the six CO₂ treatments of the RICE II experiment (Baker and Allen, 1993b; Allen et al., 1995) gave values of about 34, 50, 60, 80, 85 and 90 m mol CO₂/m²/s at high light (1600 m mol photons/m²/s) for treatments of 160, 250, 330, 500, 660 and 900 m mol/mol.

Linear canopy Pn responses to PPFD on day 60 of the RICE IV experiment were similar among all temperature treatments (25/18, 28/21, 31/24, 34/27 and 37/30°C) that were exposed to 660 m mol/mol CO₂ responses of the 330 m mol/mol treatment were about 25% less. Other studies of canopy Pn have shown only small differences across wide ranges of temperature; e.g., cotton (Baker et al., 1972) and soybean (Jones et al., 1985a). There may be two reasons for the lack of a clear canopy photosynthetic response to air temperature across the 25 to 37°C range. Evaporative cooling may lower foliage temperature below air temperature increasingly with increasing air temperature and increasing vapour pressure deficit (Allen, 1990; Pickering et al., 1995). The summed response of the photosynthetic rates of leaves to temperature at all exposures of light may broaden the temperature response for the whole canopy photo-synthetic rates (Pickering et al., 1995).

Figure 4.1. Photosynthesis CO₂ assimilation rate (A) as a function of intercellular CO₂ (C_i) for single, attached, fully expanded leaves of rice plants grown at CO₂ concentrations of 330 (open symbols) or 660 (filled symbols) m mol/mol and dry bulb air temperatures (day/night) of 35/26°C. Measurements were made on 29 October 1992 at an irradiance of 1 200 to 1 300 m mol (photons)/m²/s. Adapted from Allen et al. (1995)

Baker and Allen (1993b) fitted whole-day canopy Pn data to a rectangular hyperbola of the form of equation (4.1) for DAP 61 of the RICE II experiment. The parameters were 70.8 m mol/mol, 3.96 and -2.21, for K_m, Y_max and Y_i, respectively, with a relative Pn response ceiling of 1.75 at infinite [C]. The calculated percentage increase in response of whole-day canopy Pn at 660 vs. 330 m mol/mol was 36%.

Photosynthetic acclimation to CO₂ at the canopy level

To test for acclimation of canopy photosynthetic capacity, Baker et al. (1990b) used rice that was grown at 160, 250, 330, 500, 660 and 900 m mol/mol CO₂. For half-day periods (mornings) on 62,63 and 64 DAP, common CO₂ setpoints of 160, 330 or 660 m mol/mol were imposed on each chamber on those respective days. Within each of these short-term CO₂ exposure comparisons, Baker et al. (1990b, 1996b) and Baker and Allen (1993a) showed that the short-term canopy net photosynthetic rate decreased with increasing long-term CO₂ treatment (Figure 4.2). The relative effects of the short-term CO₂ exposure were greatest for the lowest short-term CO₂ concentration. For example, when compared at a common short-term CO₂ exposure of 160 m mol/mol, the canopy photosynthetic rate of the chamber containing the 900 m mol/mol long-term CO₂ growth treatment was only about one-third of that of the 160 m mol/mol long-term CO₂ growth treatment (Figure 4.2). However, a large part of this apparent acclimation effect may be attributed to the greater respiration rates of the plants that had been grown under elevated CO₂ (Boote et al., 1994).

Rubisco protein percentage was used as evidence of leaf acclimation to a wide range of CO₂ concentrations (Baker and Allen, 1994; Allen et al., 1995), shown in Table 4.5 for 34 DAP soybean (Campbell et al., 1988) and 75 DAP rice (Rowland-Bamford et al., 1991). For rice, rubisco activity expressed on a leaf area basis decreased by 66% across the 160 to 900 m mol/mol long-term CO₂ treatments (Rowland-Bamford et al., 1991). A major cause of this decline in rubisco activity was a 32% decrease in the amount of rubisco protein relative to other soluble protein (Rowland-Bamford et al., 1991).

Although Rowland-Bamford et al. (1991) reported decreases in both amount and activity of rubisco with increasing CO₂ in rice plants grown across the range of 160 to 900 m mol/mol, the leaf A/C. response curves gave no indication of a downward acclimation of photosynthetic capacity across the smaller range of 330 to 660 m mol/mol CO₂. One possibility would be that rubisco is not as limiting for photosynthesis under CO₂ enrichment as would be expected. More work is needed, under a range of CO₂ treatments, to explore the interaction effects of sink capacity, nitrogen nutrition, and other internal CO₂-fixation processes on photosynthetic behaviour and crop yield.

Figure 4.2. Comparison of rice canopy net photosynthetic rate (Pn) vs. long-term [CO₂] acclimation treatment for rice canopies grown at subambient (160, 250), ambient (330), and superambient (500, 660 and 900 m mol CO₂/mol air) [CO₂] treatments in 1987. The Pn estimates were obtained during a short-term [CO₂] change-over study where the [CO₂] was maintained during the morning hours in all six long-term [CO₂] treatments at 160, 330 and 660 m mol/mol on days 62, 63 and 64 days after planting, respectively. The Pn was estimated from linear regression equations of Pn vs. photosynthetic photon flux density (PPFD) with PPPD set to 1500 m mol photons/m²/s. The vertical bars represent 95% confidence intervals. Adapted from Baker et al. (1990b, 1996b) and Allen et al., (1995)

Table 4.5. For soybean, leaf blade soluble protein expressed on a leaf blade area basis and percentage rubisco protein expressed on a leaf blade soluble protein basis for 34-day-old soybean plants grown under a wide range of CO₂ concentrations. Adapted from Campbell et al. (1988) and Baker and Allen (1994). For rice, leaf nitrogen content expressed on a leaf area basis and percentage rubisco protein expressed on a leaf soluble protein basis for 75-day-old rice plants grown under a wide range of CO₂ concentrations. Adapted from Rowland-Bamford et al. (1991) and Baker and Allen (1994)

Soybean			Rice
CO2 growth concentration (m mol/mol)	Leaf soluble protein. (g/m2)	Rubisco protein (%)	CO2 growth concentration (m mol/mol)	Leaf nitrogen protein (m mol/m2)	Rubisco protein (%)
160	2.5	56	160	95	62
220	3.2	54	250	90	59
280	2.6	-	330	81	54
330	2.3	57	500	62	49
660	2.3	54	660	78	43
990	2.3	55	900	64	42

Development stages

In rice, two distinctly different rates of leaf appearance for vegetative and reproductive phases of growth were found (Baker et al., 1990, 1996; Baker and Allen, 1994), as has been observed many times before (Yoshida, 1977; Vergara, 1980). Leaf appearance rates (leaves per day) are about twice as great in the vegetative phase as in the reproductive phase.

In the 'late' planted (23 June 1987) RICE II experiment, the number of mainstem leaves at panicle initiation and the final number of mainstem leaves decreased across CO₂ treatments of 160 to 500 m mol/mol and remained similar from 500 to 900 m mol/mol. Panicle initiation and boot stage occurred about 12 days earlier in the superambient CO₂ treatments compared with the 160 m mol/mol treatment (Baker et al., 1990a; Baker and Allen, 1993b). Therefore, plant developmental rate was clearly accelerated with increasing CO₂ up to about 500 m mol/mol.

Temperature effect data were assembled from treatments ranging from 25/18/21 to 40/33/37°C (day/night/paddy water temperatures) in RICE I to RICE V experiments. Air temperature greatly influenced leaf appearance rate (Figure 4.3-A), developmental rate, and total growth duration, whereas, across the limited range of 330 to 660 m mol/mol the effects were comparatively small (Baker and Allen, 1993b; Baker et al., 1992a,b, 1995). No consistent differences in phyllochron interval (days per leaf appearance) between 330 and 660 m mol/mol were observed, whereas phyllochron interval increased with increasing temperature across the range of 25/18/21 to 40/33/37°C, especially for the reproductive phase (Figure 4.3-B). In RICE III, IV and V experiments, anthesis ranged from 0 to 6 days earlier in the 660 vs. 330 m mol/mol treatments (Baker et al., 1995). Also, the number of days to anthesis was shortened by approximately 10 days across the temperature treatment range from 25/18/21 to 34/27/31°C for the RICE IV experiment.

Figure 4.3. (A) Mainstem Haun scale growth units vs. days after planting for rice plants in two different CO₂ and temperature regimes from the RICE III experiment.

Figure 4.3. (B) Phyllochron interval vs. temperature treatment for all five rice experiments. 'Vegetative' data are plotted against paddy water temperature and 'reproductive' data are plotted against average day/night temperature adjusted for thermoperiod.

Adapted from Baker et al. (1995) and Allen et al. (1995)

Growth and yield

The yield and yield component data of the RICE I and RICE II experiments with CO₂ concentrations spanning 160 to 900 m mol/mol are given in Table 4.6. A rectangular hyperbola (equation (4.1)) was fitted to the seed yield data. Values of K = 284 m mol/mol, Y_max =2.24 and Y_i= -0.13 were obtained for relative grain yield with asymptotic relative yield response ceiling (Y_max + Y_i) = 2.11 at infinite [C]. The calculated percentage increase in response at 660 vs. 330 m mol/mol was 44% for grain yield (Baker and Allen, 1993b).

Across a relatively wide temperature range from 25/18/21 to 37/30/34°C, Baker et al. (1992b) found a broad temperature optimum for biomass production in the mid-temperature ranges. Plants grown at 40/33/37°C were near the upper temperature limit for survival. High temperature spikelet sterility of rice is induced almost exclusively on the day of anthesis (Satake and Yoshida, 1978) when temperatures greater than 35 °C for more than one hour induce a high percentage of sterility (Yoshida, 1981). In the 40/33/37°C treatments, plants in the 330 m mol/mol CO₂ chamber died during internode elongation whereas plants in the 660 m mol/mol chamber produced small, abnormally shaped panicles that were sterile (Baker et al., 1992a). Therefore, elevated CO₂ may slightly increase the maximum temperature at which rice plants can survive.

At both CO₂ levels, grain yield was highest in the 28/21/25°C treatment followed by a decline to zero yield in the 40/33/37°C treatment (Table 4.7). The CO₂ enrichment from 330 to 660 m mol/mol increased yield by increasing the number of panicles per plant, whereas the number of filled grains per panicle and individual seed mass were less affected. Temperature effects on yield and yield components were highly significant. The number of panicles per plant increased while the number of filled grains per panicle decreased sharply with increasing temperature treatment. Individual seed mass was stable at moderate temperatures but tended to decline at temperature treatments above 34/27/31°C. Final above-ground biomass and harvest index were increased by CO₂ enrichment while harvest index declined sharply with increasing temperature. Notably, there were no significant CO₂ x temperature interaction effects on yield, yield components, or final above-ground biomass (Table 4.7).

At each CO₂ concentration, polynomial regression equations were fitted to the rice seed yield (Y) vs. temperature (X) data of Table 4.7. A third-degree polynomial [Y = -239.226 + 25.557 * X - 0.848 * X² + 0.00896 * X³ (R²= 0.91, S_y.x= 1.36)] provided the best fit for the 660 m mol/mol CO₂ treatments (Baker and Allen, 1993a). However, a second-degree polynomial [Y = -1.196 + 1.042 * X - 0.0278 * X² (R² = 0.91, S = 1.36)] fits the data best for the 330 m mol/mol treatments. Figures 4.4 and 4.5 show the effect of temperature on seed yield and final biomass, respectively, from a number of experiments based on the tables and figures shown in Baker and Allen (1993a). While future increases in atmospheric CO₂ should benefit rice yields, large negative effects are likely if temperatures also rise. The figures show that vegetative productivity is maintained at higher temperatures than is reproductive growth.

Table 4.6. Grain yield, components of yield, total above-ground biomass and harvest index of rice subambient and superambient CO₂ concentration experiments conducted in 1987. Adapted from Baker et al. (1988), Baker and Allen (1993b, 1994), Baker et al. (1995,1996).

CO2	Temperature 1	Grain yield	Panicle/plant	Filled grain	Grain mass	Biomass	Harvest index	Experiment RICE No.
m mol/mol	°C	Mg/ha	no./plant	no./panicle	mg/seed	g/plant
160	31/31/27	3.4c2	3.6c	24.8a	17.0a	4.0c	0.36a	I & II
250	31/31/27	4.1c	4.8bc	20.8a	18.2a	5.1bc	0.34a	I & II
330	31/31/27	4.8bc	5.7ab	21.0a	17.6a	6.3ab	0.34a	I & II
500	31/31/27	6.83	7.3	23.0	18.1	9.8	0.30	II only
660	31/31/27	6.6ab	6.5ab	25.0a	17.8a	8.4a	0.35a4	I & II
900	31/31/27	7.3a	7.4a4	24.8a	17.8a	8.2a	0.39a	I & II

¹ Daytime dry bulb air temperature/nighttime dry bulb air temperature/paddy water temperature.

² Values followed by the same letter in the same column are not significantly different by Duncan's Multiple Range Test (p - 0.05).

³ Plants in 500 m mol/mol CO₂ treatment of the RICE I experiment experienced hydrogen sulphide toxicity in the root zone which inhibited seed fill and reduced grain yield to levels below those measured in subambient treatments. These data were not used in this analysis. The values in the 500 m mol/mol CO₂ treatment row are from the RICE II experiment only.

⁴ Indicates significant difference between planting dates as determined by the t-test at the 0.05 level of confidence.

Table 4.7. Grain yield, components of yield, total above-ground biomass and harvest index for five separate rice experiments. Adapted from Baker and Allen (1993a) and Baker et al. (1996)

CO2	Temperature 1	Mean air temperature 2	Grain yield	Panicle/plant	Filled grain	Grain mass	Biomass	Harvest index	Experiment No.
m mol/mol	°C	°C	Mg/ha	no./plant	no./panicle	mg/seed	g/plant
330	28/21/25	24.2	7.9	5.1	34.5	17.4	7.3	0.47	RICE III
	28/21/25	25.1	6.6	3.9	39.6	17.5	6.5	0.44	RICE IV
	28/21/25	25.1	8.0	4.0	47.5	18.5	8.1	0.43	RICE V
	31/31/27	31.0	5.2	5.9	23.0	17.1	5.5	0.42	RICE I
	31/31/27	31.0	4.3	5.4	19.0	17.9	7.2	0.26	RICE II
	34/27/31	30.2	4.2	7.7	15.2	16.2	5.6	0.43	RICE III
	40/33/37	36.2	0.0	-.-	-.-	-.-	-.-	-.-	RICE III
660	25/18/21	22.1	8.4	4.4	46.0	18.2	7.8	0.47	RICE IV
	28/21/25	24.2	8.4	5.0	37.7	18.0	7.9	0.46	RICE III
	28/21/25	25.1	10.4	4.2	58.3	18.3	8.9	0.50	RICE IV
	28/21/25	25.1	10.1	4.4	54.2	19.0	9.3	0.47	RICE V
	31/31/27	31.0	6.8	6.9	25.1	17.2	7.5	0.40	RICE I
	31/31/27	31.0	6.4	6.0	24.8	18.4	9.3	0.29	RICE II
	34/27/31	30.2	4.8	6.5	18.5	16.7	6.3	0.32	RICE III
	34/27/31	31.1	3.4	7.5	12.9	16.3	8.1	0.18	RICE IV
	37/30/34	34.1	1.0	8.0	3.0	14.2	7.1	0.06	RICE IV
	40/33/37	36.2	0.0	-.-	-.-	-.-	-.-	-.-	RICE III
F-values
CO2			4.2*	4.3*	0.4 NS	0.1 NS	13.1 **	4,6*
Temperature			51.8**	22.6 **	51.1 **	27.0 **	3.2*	32,8 **
CO2 x temperature			1.9 NS	1.7 NS	1.3 NS	0.3 NS	0.2 NS	1.9 NS

¹ Daytime dry bulb air temperature/nighttime dry bulb air temperature/paddy water temperature.

² Mean air temperature is the average of day and night temperature adjusted for thermoperiod.

** * Significant at the 0.01 and 0.05 probability levels, respectively. NS = Not significant.

Figure 4.4. Rice seed yield vs. weighted mean day/night air temperature for plants grown to maturity in CO₂ concentrations of 330 and 660 m mol/mol in five separate experiments

Figure 4.5. Second-degree polynomial fit of rice final biomass yield vs. weighted mean day/night air temperature for plants grown to maturity in CO₂ concentrations of 330 and 660 m mol/mol in five separate experiments

Canopy dark respiration rates

Plant respiration rates may be decreased by both short- and long-term exposure to high CO₂ concentrations (Bunce, 1990; Amthor, 1991); however Baker et al. (1992c) found that nighttime canopy dark respiration rates [R_d, m mol (CO₂)/m² (ground area)/s] increased for rice exposed to daytime CO₂ ranging from 160 to 900 m mol/mol. Similar to photosynthetic rates, R_d increased with CO₂ exposure from 160 to 500 m mol/mol but levelled off somewhat across the 500 to 900 m mol/mol range. The R_d of the ambient and superambient CO₂ treatments reached a broad maximum around 30 to 50 DAP whereas the broad maximum of the subambient treatments occurred later around 50 to 70 DAP. The maximum values of R_d were about 6, 8, and 9 m mol/m²/s for the 160, 250 and 330 m mol/mol CO₂ treatments and about 11 to 12 m mol/m²/s for the three superambient CO₂ treatments.

Specific respiration rate [R_dw , m mol (CO₂)/s/kg (total above-ground dry matter)] decreased exponentially with DAP at all CO₂ exposures and was higher in the subambient (160 and 250 m mol/mol) than in the ambient (330 m mol/mol) and superambient (500, 660 and 900 m mol/mol treatments). At each CO₂ exposure level, the patterns of R_dw with DAP were very similar to the patterns of plant tissue nitrogen concentration with DAP (Baker et al., 1992c). Furthermore, R_dw was linearly related to total above-ground plant tissue nitrogen concentration [mg (N)/g (DW)] across the range of CO₂ exposures and the six dates of plant sampling (R_dw = -13.0 + 0.952 * [N]; r2=0.9, P=0.01). Baker et al. (1992c) concluded that the CO₂ treatments affected R_dw mainly by altering the protein content of the plant tissue. Another explanation is that elevated CO₂ increased the amount of structural and non-structural carbohydrates in the plant tissues, so that a larger proportion of dry matter was sequestered in non-protein materials. This study showed no respiration acclimation to long-term CO₂ enrichment (indirect acclimation) of rice that could not be explained by nitrogen concentration of the plant tissues.

Evapotranspiration and water-use efficiency

Both plant transpiration and direct evaporation from the floodwater surface contribute to water use (evapotranspiration, ET) of rice growing in the SPAR chambers. Diurnal trends of ET followed diurnal patterns of solar irradiance (Baker et al., 1990b; Allen et al., 1995). After canopy closure, maximum ET rates and total daytime water losses were about 35 and 30% greater, respectively, from rice grown at 160 m mol/mol compared to rice grown at 900 m mol/mol CO₂ (Table 4.8). The ET rates were similar when all chambers were exposed for one-half day to the same CO₂ concentration (data not shown, Baker et al., 1990b), which demonstrates the effect of the exposure level of CO₂ on stomatal control of transpiration.

Temperature also has a large effect on diurnal ET rates (Baker and Allen, 1993a; Allen et al., 1995) and daytime water use (Table 4.8), mediated primarily through vapour pressure deficit of the air. Solar irradiance also has a large effect, probably through both the energy inputs to the canopy and through the stomatal opening response to light (directly or indirectly). Midday maximum ET rates were about 75 and 35% higher for rice grown at 40/33/37°C and 34/27/31°C, respectively, compared to rice grown at 28/21/25°C (Baker and Allen, 1993a; Allen et al., 1995).

Water-use efficiency: effects of CO₂

Daytime totals of CO₂ uptake, ET, and calculations of WUE vs. CO₂ treatment are shown in Table 4.8. Stomatal conductance decreases with increasing CO₂ concentration which can cause a reduction of both leaf and whole canopy transpiration. However, CO₂ enrichment may also increase canopy leaf surface area for transpiration, thereby offsetting some of the water savings (Jones et al., 1985b; Allen et al., 1985). The leaf area index of the RICE II experiment ranged from 7.6 to 10.8 across the CO₂ treatments from 160 to 900 m mol/mol. Calculations of WUE increased with increasing CO₂ (Table 4.8) due to the decline in ET across this CO₂ range and the increase in Pn with CO₂ up to the 500 m mol/mol treatment.

Table 4.8. Comparison of total daytime responses of rice canopies grown season-long in various CO₂ concentration (RICE II, 23 August 1987, 61 days after planting) and temperature (RICE IV, 10 September 1989, 58 days after planting) treatment regimes. Daytime dry bulb air temperature/nighttime dry bulb air temperature/paddy water temperature: dewpoint temperature treatments are given in the second column. Adapted from Baker et al. (1990b) and Baker and Allen (1993a,b, 1994).

CO2	Temperature 1	Total daytime photons	Total daytime CO2 uptake	Total daytime H2O loss	Water-use efficiency 2
m mol/mol	°C	mol(Photons)/m2	mol(CO2)/m2	mol(H2O)/m2	mmol/mol
	Effects of carbon dioxide: RICE II, 23 August 1987, 61 DAP
160	31/31/27:18	39.4	0.74	608.5	1.22
250	31/31/27:18		1.13	643.4	1.76
330	31/31/27:18		1.34	611.0	2.19
500	31/31/27:18		1.80	553.6	3.25
660	31/31/27:18		1.79	536.2	3.34
900	31/31/27:18		1.83	469.5	3.90
	Effects of temperature: RICE IV, 10 September 1989, 58 DAP
330	28/21/25:12.0	26.4	0.87	612.7	1.42
660	25/18/21:10.5		0.83	359.3	2.30
660	28/21/25:12.0		1.13	475.8	2.37
660	31/24/28:13.5		0.83	491.3	1.69
660	34/27/31:15.0		0.89	736.4	1.20
660	37/30/34:16.5		0.97	909.4	1.06

¹ Daytime dry bulb air temperature/nighttime dry bulb air temperature/paddy water temperature: dewpoint temperature.

² mmol CO₂/mol (H₂O).

Water-use efficiency: effects of temperature

In the 28°C air temperature treatment, CO₂ enrichment from 330 to 660 m mol/mol resulted in 30% increase in daytime total CO₂ uptake and 22% decrease in ET with a concomitant increase of WUE from 1.42 to 2.37 m mol (CO₂)/mmol (H₂O), an increase of 67% (Table 4.8). Daytime ET was more than doubled from the 25 to the 37°C daytime temperature treatment. WUE declined from the 28 to the 37°C daytime treatment due to the sharp increase in ET and the relatively stable Pn across this temperature range. If WUE were based on grain yield, it would decrease drastically with increasing temperature not only because of increasing ET, but also because of sharply decreasing seed production.

Rising atmospheric CO₂ is likely to benefit rice production by increasing photosynthesis, growth and grain yield while reducing water use and increasing WUE. In warm areas of the world, however, possible future global warming may result in substantial yield decreases because of the sensitivity of flowering and seed set to high temperatures and the possibility of water shortages that may result from increased evapotranspiration.

Summary of comprehensive reviews

Several recent symposia proceedings and reviews leave little doubt that crop plants can respond well to elevated CO₂ (Rozema et al., 1993; Woodwell and Mackenzie, 1995; Wittwer, 1995). Poorter (1993) compiled information from 156 plant species and found that doubling CO₂ provided an average growth increase of 37%. The distribution of weight ratios of CO₂-enriched and control plants is shown in Figure 4.6. Poorter's compilation showed a 41 and a 22% increase for C₃ and C₄ plants, respectively. As a group, C₃ herbaceous crop plants responded more than wild herbaceous species (58 vs. 35%). Furthermore, the fast growing wild species responded more strongly than slow-growing wild species (54 vs. 23%).

Poorter (1993) imposed two restrictions on his compilations that may have led to larger than expected responses to elevated CO₂. Firstly, plants grown in competition were not included. Secondly, only vegetative stages of plants were compared since compiled data were selected prior to flowering.

A number of studies have shown that vegetative growth responses may be greater than reproductive (seed yield) responses. Therefore, the compiled data of Poorter may give an impression of greater response than would be observed throughout the life cycle. Secondly, crops in field conditions usually are grown in dense populations where they compete for space and light. Under more realistic field conditions, crop plants are likely to respond as a community rather than individual plants, wherein light (solar radiation) becomes a limiting factor for growth. Under these conditions, elevated CO₂ cannot promote horizontal expansion and greater light capture. Although the actual field responses may be less, the CO₂ fertilization effect is clearly well-established.

Figure 4.6. Distribution of the biomass growth modification ratio (weight ratio) of CO₂-enriched plants (600 to 720 m mol/mol) in comparison with control treatments (300 to 360 m mol/mol). The bar graph was created from averages of all the weight ratios of 156 species selected from the literature. Source: Poorter (1993)

The CO₂ fertilization effect for forest species has also become firmly established. Wullschleger et al. (1994) estimated the biotic growth factor for 58 controller-exposure studies of forest tree species which included 398 observations. Their frequency distribution of relative growth response of trees grown at elevated CO₂ vs. ambient CO₂ is given in Figure 4.7. They also found a mean response ratio of 1.32. Of the 398 observations, 51 showed a relative growth response less than 1, and 31 showed responses greater than 2. However, under competitive conditions, tree response may be much less (Bazzaz et al., 1995).

Kimball et al. (1993) discussed the data of Idso et al. (1987) which related the growth modification factor (or biomass growth modification ratio) caused by a 300 m mol/mol increase in CO₂ concentration above ambient. Their data are reproduced as Figure 4.8. Under Phoenix, Arizona, conditions, the growth modification factor across a mean temperature range of 12 to 34 °C was rather linear. Thus, the vegetative response to CO₂ should be enhanced at increased temperatures. As discussed before, this growth modification factor may not apply for reproductive growth responses (seed yield) of crops (e.g., rice).

Figure 4.7. Frequency distribution of the log-transformed biomass growth modification ratio (log elevated/ambient CO₂ exposure total dry mass ratios) of 73 tree species. Source: Wullschleger et al. (1994)

Crop modelling: Predictions for the future

MODELLING CROP RESPONSES TO CO₂ AND CLIMATE CHANGES

Peart et al. (1989) and Curry et al. (1990a,b, 1995) predicted growth and yield responses of soybean and maize to doubled-CO₂ climate change scenarios of the southeastern USA. Their simulations used 30 years of baseline weather data (1951-1980) from 19 sites in 11 states. Predicted climate changes of the Goddard Institute for Space Studies (GISS) model (Hansen et al., 1988) and the Geophysical Fluid Dynamics Laboratory (GFDL) model (Manabe and Wetherald, 1987) were used to change temperatures, precipitation and solar radiation, month-by-month, for the 30 years of baseline data at each site. These modified baseline weather data provided GISS and GFDL climatic change scenarios (Smith and Tirpak, 1989). At each site, prevailing cropped soil types, planting dates, and cultivars were used in the simulations.

Figure 4.8. Biomass growth modification ratio (growth modification factor or relative increase in biomass growth) resulting from a 300 m mol/mol increase in CO₂ concentration above ambient (almost doubled) versus mean daily air temperature for the plants indicated in the legend. Source: Kimball et al. (1993)

Baseline, GISS and GFDL scenario temperature and rainfall data that were used by Peart et al. (1989) were condensed for two representative locations; Columbia, South Carolina, and Memphis, Tennessee (Table 4.9). Crop yield responses to climate change were simulated under four conditions: with or without direct CO₂ fertilization effects, and under rainfed or optimum irrigation culture. The Penman-Monteith equation, which contains a term for canopy conductance, was used to compute the effects of elevated CO₂ on canopy transpiration. Simulations were run for doubled-CO₂ conditions with a crop photosynthetic enhancement factor of 1.35 for soybean (a C₃ plant) and 1.10 for maize (a C₄ plant). No simulations were run with CO₂ fertilization effects only (without climate change effects).

Simulated soybean seed yields were averaged over 30 years and 19 sites (Table 4.10). Under rainfed conditions with climate change effects only, simulated soybean yields under the GFDL scenario were reduced 71 % compared to the baseline climate, whereas the average yields were reduced only 23% under the GISS scenario. Yields under the GFDL scenario were severely impacted because of the rainfall reductions predicted by this model (Table 4.9).

Table 4.9. Summary of 30-year mean (1951-1980) baseline (BASE) weather data of two locations with GISS and GFDL scenarios. Condensed from Peart et al. (1989)

	BASE			GISS			GFDL
	Prec. (mm)	Tmax (°C)	Tmin (°C)	Prec. (mm)	Tmax (°C)	Tmin (°C)	Prec. (mm)	Tmax (°C)	Tmin (°C)
Columbia. SC
APR-SEP; Jul	684	33.3	21.2	817	35.2	23.1	471	38.2	26.1
OCT-MAR; Jan	561	13.5	0.7	602	15.8	3.0	571	15.6	2.8
TOTAL; Mean	1245	24.1	10.7	1419	26.6	13.2	1042	27.3	13.9
Memphis. TN
APR-SEP; Jul	656	33.1	22.6	748	35.4	24.9	549	36.0	25.5
OCT-MAR; Jan	654	9.0	-0.6	565	12.0	2.4	758	11.3	1.7
TOTAL; Mean	1 310	22.0	11.1	1 313	25.4	14.5	1307	24.8	13.9

Under optimum irrigation conditions, average yields under both the GISS and GFDL scenarios were reduced 18 to 19%. However, in spite of higher temperatures, the irrigated yields under the GISS and GFDL scenarios were about 25% greater than the baseline climate scenario without irrigation.

When CO₂ fertilization effects were included with climate change effects (Table 4.10), simulated average yields of the GISS climate scenario were increased 11% under rainfed conditions, whereas yields under the GFDL climate scenario were still decreased (-52%). Under optimum irrigation with CO₂ fertilization effects, yields under both GISS and GFDL scenarios were increased 13 to 14%.

Yields of maize were simulated for doubled-CO₂ fertilization effects at only four weather station locations (Charlotte, North Carolina; Macon, Georgia; Memphis, Tennessee; Meridian, Mississippi). For climatic change effects alone, predicted maize yields declined only 6% in the GISS scenario, but declined 73% in the GFDL scenario (Table 4.11). Although irrigation increased predicted crop yields, the GISS and GFDL climate scenarios gave yield decreases of 18 and 27%, respectively, with respect to the irrigated baseline weather. The yield reduction of the GFDL scenario with respect to the GISS scenario was 10%, attributable to slightly higher temperatures. Including the CO₂ fertilization effects with climatic change scenarios had little effect on the predicted yields of maize because it is a C₄ plant.

In the Great Lakes and Corn Belt Area, simulations by Ritchie et al. (1989) showed that higher temperature would have the greatest effect of the climate change factors on predicted yields of soybean and maize, mainly through decreases in the crop life cycle. For the southern part of this region, yield reductions were greater for the GFDL scenario. Predicted yields increased for the northernmost stations because temperatures and growing season duration became more favourable. Average irrigation water requirements in this region increased about 90% for the GISS and GFDL scenarios.

Rosenzweig (1989) modelled maize and wheat yields in the Great Plains under GISS and GFDL scenarios. Yields decreased in the Southern Great Plains because higher temperatures shortened the life cycle of the crops. Where precipitation was predicted to decrease, irrigation requirements increased. In a separate modelling study, Allen and Gichuki (1989) predicted a 15% increased requirement for irrigation for this region, with greater requirements for alfalfa because its growing season was increased, and lower requirements for maize and winter wheat because their growing seasons were decreased. The CO₂ fertilization effect offset the adverse effects of climate change at some locations of Ritchie et al. (1989) and Rosenzweig (1989).

Table 4.10. Doubled CO₂ soybean yield simulations (SOYGRO) for the southeastern USA

BASE	GISS Model		GFDL Model		Model
Yield (kg/ha)	Yield (kg/ha)	% Diff.	Yield (kg/ha)	% Diff.	% Diff.
Climate Change Effects Only, Rainfed
2497	1 929	-23	733	-71	-62
Climate Change Effects Only, Irrigated
3 837	3 158	-18	3092	-19	- 2
CO2 Fertilization plus Climate Change Effects, Rainfed
2497	2780	+11	1 206	-52	-57
CO2 Fertilization plus Climate Change Effects. Irrigated
3837	4350	+13	4393	+14	+1

Adapted from Peart et al. (1989).

Table 4.11. Doubled CO₂ maize yield simulations (CERES-Maize) for Charlotte NC, Macon GA, Meridian MS, and Memphis TN

BASE	GISS Model		GFDL Model		Model
Yield (kg/ha)	Yield (kg/ha)	% Diff.	Yield (kg/ha)	% Diff.	% Diff.
Climate Change Effects Only, Rainfed
8468	7926	-6	2289	-73	-71
Climate Change Effects Only. Irrigated
13,899	11 455	-18	10257	-26	-10
CO2 Fertilization plus Climate Change Effects. Rainfed
8577	8 136	-5	2224	-74	-73
CO2 Fertilization plus Climate Change Effects. Irrigated
14052	11 545	-18	10363	-26	-10

Adapted from Peart et al. (1989).

Dudek (1989) predicted productivity changes of several Californian vegetable, fruit and nut crops in response to GISS and GFDL scenarios for doubled CO₂. Without CO₂ fertilization effects, statewide average yield changes, depending on the crop, would be -8 to -34% for the GISS scenario and -6 to -31 % for the GFDL scenario. With CO₂ fertilization effects, predicted statewide yield changes ranged from about +17 to -12% for the GISS scenario and about +21 to -8% for the GFDL scenario.

Dudek (1989) used a California Agriculture and Resource Model (CARM) to further predict economic and market impact of the productivity changes. Production declined, in general, under the scenarios of climate change without CO₂ fertilization effects; however, commodity prices generally increased. Under the CO₂ fertilization plus climate change effects scenarios, predicted impacts on commodity prices were much less.

The effect of temperature on the phenology of crop plants plays a critical role. One crucial need is for more detailed research on responses of plants to temperature, and temperature x CO₂ interactions, as inputs to crop models. More modelling studies are also needed on different planting dates as an adaptive strategy. Cultivars need to be designed for future climatic conditions (Hall and Allen, 1993). Factors that should be considered are: extension of the grain-filling period and perhaps shortening the duration of vegetative growth (which would also improve harvest index); ability to flower and set seed at higher temperatures; photoperiod and thermoperiod adaptive interactions; selection for positive photosynthetic acclimation where negative photosynthetic acclimation has been observed; and capability of utilizing photoassimilates (storage carbohydrates) more effectively. These factors need to be integrated into whole-plant physiology under real-world conditions.

ADAPTATIONS AND EVAPOTRANSPIRATION REQUIREMENTS

The simulated crop yield responses to climatic changes provided by Peart et al. (1989) and Curry et al. (1990a,b, 1995) manifest two main points: (a) the serious adverse impact of inadequate rainfall scenarios on crop production coupled with rising temperature scenarios, and (b) the importance of beneficial CO₂ fertilization effects in the face of elevated temperatures. However, the climate change for an effective doubling of CO₂ may occur at CO₂ concentrations less than those used in this simulation, if radiatively active trace gases other than CO₂ play a large role in the greenhouse effect. In that case, the direct CO₂ effects would be somewhat lower than shown in the example of Tables 4.10 and 4.11 for an equivalent climate change. All of the simulations assumed that climatic changes would occur simultaneously with increasing concentrations of CO₂ and other trace greenhouse-effect gases. If global warming lags the increases of atmospheric CO₂, then some beneficial effects of CO₂ fertilization are likely to occur before the full impact of climate change is manifested. However, Broecker (1987) and others caution that climate changes have not always been gradual during interglacial periods of the Pleistocene Era. There is clear evidence of relatively rapid climatic oscillations in the northern hemisphere during the previous interglacial period 110000 to 140000 years before present based on Greenland Ice-core Project (GRIP) records (Anklin et al., 1993). These oscillations produced cold periods that were as severe as the proceeding glacial period.

Much of the reduction in soybean yields reported by Peart et al. (1989) and Curry et al. (1990a,b, 1995) was caused by decreases in the length of the grain-filling period under higher temperatures. Changes in management practices, such as changing planting dates or selection of other cultivars, may help to prevent some of the potential reductions in yield. In the future, plant breeders may need to adapt combinations of temperature tolerance and photoperiod responses into new germplasm. In situations where non-structural carbohydrates accumulate as a CO₂ fertilization effect response, new germplasm needs to be developed that can make better use of the photoassimilate source.

Irrigation is not likely to be a panacea for climate change. Predicted irrigation requirements for soybean in the southeastern USA were increased 33 and 134% under the GISS and GFDL scenarios, respectively, based on simulations of Peart et al. (1989) and Curry et al. (1990a,b). However, under the GFDL scenario, water resources would become scarce, and may not be readily available for crops. Some areas of the USA may have to adapt by irrigating less land area. Increasing temperatures and decreasing precipitation for the USA as predicted by the GFDL model would have a serious negative impact overall on agricultural productivity although producers in favourable regions may benefit from scarcity-mediated higher prices (Adams et al., 1990).

ASSESSING INTERNATIONAL IMPACTS OF RISING CO₂ AND CLIMATE CHANGE

Several assessments have been conducted on the impacts of rising CO₂ and global climatic changes on crop production patterns and economics responses within national or regional zones (Adams et al., 1990; Crosson, 1993; Parry et al., 1988; Smith and Tirpak, 1989) and in various countries around the world (Rosenzweig and Iglesias, 1994; Rosenzweig et al., 1995). Furthermore, progress has been made on predicting impacts of climate change using world trade models (Rosenzweig and Parry, 1993, 1994).

The studies conducted by Rosenzweig and colleagues used existing crop models and climate change scenarios from three GCMs (GISS, GFDL and United Kingdom Meteorological Office Model (UKMO), Wilson and Mitchell, 1987) to predict changes in crop yields for a number of countries around the world. The doubled-CO₂ climate change scenarios (temperature, rainfall and evaporation changes) were based on the climate change potential expected from increases of all greenhouse-effect gases. These climate changes are expected to occur well before CO₂ concentration has actually doubled. Therefore, the CO₂ levels for fertilization effects were estimated to be 555 m mol/mol rather than 660 m mol/mol. Thus, the CO₂ fertilization effect photosynthetic ratios for four crops (soybean, wheat, rice and maize) were taken as 1.21, 1.17, 1.17 and 1.06, respectively.

The impact of climate change scenarios was more severe in the tropical latitudes than in the mid- or high-latitudes. For example, averaged over all three GCM scenarios, wheat production changes predicted in Brazil, India, China and Canada were -47, -43, -11 and -20%, respectively, without CO₂ effects, and -28, -13, +8 and +16% with CO₂ fertilization effects (calculated from data of Rosenzweig and Parry, 1993; Rosenzweig and Iglesias, 1994).

Simulations of soybean yields throughout the USA were a part of this international study (Curry et al., 1995). They found that aggregated yields were 2.42, 2.80, 2.37 and 1.31 Mg/ha for the BASE, GISS, GFDL and UKMO scenarios, respectively. Temperatures were about 4.0°C higher than BASE for the GISS and GFDL scenarios, but were about 5.2°C higher for the UKMO model.

Refinements and improvements in prediction methodology will continue, but these assessments provide the best currently available insights into the CO₂ fertilization and climate change effects on global crop productivity.

Summary and conclusions

Elevated CO₂ increases the size and dry weight of most C₃ plants and plant components. Relatively more photoassimilate is partitioned into structural components (stems and petioles) during vegetative development in order to support the light-harvesting apparatus (leaves). The harvest index tends to decrease with increasing CO₂ concentration and temperature. Selection of plants that could partition more photoassimilates to reproductive growth should be a goal for future research. As more is learned about the effects of anticipated climate changes on crops, more effort should be directed to exploring biological adaptations and management systems for reducing these impacts on agriculture and humanity. Whether regional climates become drier or wetter with global warming remains to be seen.

Acknowledgements

Supported in part by the US Department of Energy Interagency Agreements DE-AI02-93ER61720, DE-AI05-88ER69014, and DE-AI01-81ER60001; and by US EPA Interagency Agreement DW 12934099 with the US 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 No. R-00000.

References

Acock, B. and Allen, L.H. Jr. 1985. Crop responses to elevated carbon dioxide concentration. In: Direct Effects of Increasing Carbon Dioxide on Vegetation. DOE/ER-0238. B.R. Strain and J.D. Cure (eds.). US Dept. of Energy, Carbon Dioxide Res. Div., Washington DC. pp. 53-97.

Adams, R.M., Rosenzweig, C., Peart, R.M., Ritchie, J.T., McCarl, B.A., Glyer, J.D., Curry, R.B., Jones, J.W., Boote, K.J. and Allen, L.H. Jr. 1990. Global climate change and US 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. 1991. Effects of increasing carbon dioxide levels and climate change on plant growth, evapotranspiration, and water resources. In: Proceedings of a Colloquium on Managing Water Resources in the West Under Conditions of Climatic Uncertainty. 14-16 Nov. 1990, Scottsdale, AZ. National Research Council, National Academy Press, Washington DC. pp. 101-147.

Allen, L.H. Jr. 1994. Carbon dioxide increase: Direct impacts on crops and indirect effects mediated through anticipated climatic changes. In: Physiology and Determination of Crop Yield. K.J. Boote, J.M. Bennett, T.R. Sinclair and G.M. Paulsen (eds.). American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America, Madison, Wisconsin. pp. 425-459.

Allen, L.H. Jr. and Amthor, J.S. 1995. Plant physiological responses to elevated CO₂, temperature, air pollution, and UV-B radiation. In: Biotic Feedbacks in the Global Climatic System: Will the Warming Increase the Warming? G.M. Woodwell and F.T. Mackenzie (eds.). Oxford University Press, New York. pp. 51-84.

Allen, R.G. and Gichuki, F.N. 1989. Effects of projected CO₂-induced climatic changes on irrigation water requirements in the Great Plains States (Texas, Oklahoma, Kansas, and Nebraska). In: The Potential Effects of Global Climate Change on the United States. Appendix C, Agriculture, Vol. 1. EPA-230-05-89-053. J.B. Smith and D.A. Tirpak (eds.). US Environmental Protection Agency, Washington DC. pp. 6-1 to 6-42.

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

Allen, L.H. Jr., Boote, K.J., Jones, J.W, Jones, P.H., Valle, R.R., Acock, B., Rogers, H.H. and Dahlman, R.C. 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., Vu, J.C.V., Valle, R.R., Boote, K.J. and Jones, P.H. 1988. Nonstructural carbohydrates and nitrogen of soybean grown under carbon dioxide enrichment. Crop Sci. 28: 84-94.

Allen, L.H. Jr., Valle, R.R., Mishoe, J.W, Jones, J.W. and Jones, P.H. 1990. Soybean leaf gas exchange responses to CO₂ enrichment. Soil Crop Sci. Soc. Fla. Proc. 49: 192-198.

Allen, L.H. Jr., Bisbal, E.C., Boote, K.J. and Jones, P.H. 1991. Soybean dry matter allocation under subambient and superambient levels of carbon dioxide. Agron. J. 83: 875-883.

Allen, L.H. Jr., Baker, J.T., Albrecht, S.L., Boote, K.J., Pan, D. and Vu, J.C.V. 1995. Carbon dioxide and temperature effects on rice. In: Climate Change and Rice. S. Peng, K.T. Ingram, H.-U. Neue, and L.H. Ziska (eds.). Springer-Verlag, Berlin, Heidelberg. pp. 256-277.

Amthor, J.S. 1991. Respiration in a future, higher-CO₂ world. Plant Cell Environ. 14: 13-20.

Amthor, J.S. 1995. Plant respiration responses to elevated partial CO₂ pressures. In: Advances in Carbon Dioxide Effects Research. L.H. Allen, Jr., M.B. Kirkham, C. Whitman and D.M. Olzyk (eds.). Special Pub., American Society of Agronomy, Madison, Wisconsin.

Anklin, M., and thirty-nine other Greenland Ice-core Project (GRIP) members. 1993. Climate instability during the last interglacial period recorded in the GRIP ice core. Nature 364: 203-207.

Arp, W.J. 1991. Effects of source-sink relations on photosynthetic acclimation to elevated CO₂. Plant Cell Environ. 14: 869-875.

Baker, J.T. and Allen, L.H. Jr. 1993a. Contrasting crop species responses to CO₂ and temperature: Rice, soybean, and citrus. Vegetatio 104/105: 239-260. Also: pp. 239-260. In: CO₂ and Biosphere. (Advances in Vegetation Science 14). J. Rozema, H. Lambers, S.C. van de Geijn and M.L. Cambridge (eds.). Kluwer Academic Publishers, Dordrecht.

Baker, J.T. and Allen, L.H. Jr. 1993b. Effects of CO₂ and temperature on rice: A summary of five growing seasons. J. Agric. Meteorol. (Japan) 48: 575-582.

Baker, J.T. and Allen, L.H. Jr. 1994. Assessment of the impact of rising carbon dioxide and other potential climate changes on vegetation. Environ. Pollut. 83: 223-235.

Baker, D.N. and Enoch, H.Z. 1983. Plant growth and development. In: CO₂ and Plants. E.R. Lemon (ed.). Westview Press, Boulder, Colorado. pp. 107-130.

Baker, D.N., Hesketh, J.D. and Duncan, W.G. 1972. Simulation of growth and yield in cotton: I. Gross photosynthesis, respiration, and growth. Crop Sci. 12: 431-435.

Baker, J.T., Allen, L.H. Jr., Boote, K.J., Rowland-Bamford, A.J., Jones, J.W, Jones, P.H., Bowes, G. and Albrecht, S.L. 1988. Response of vegetation to carbon dioxide, Ser. No. 043. Response of rice to subambient and superambient carbon dioxide concentrations, 1986-1987. Report of the US Dept. of Agriculture, Agricultural Research Service (in cooperation with the University of Florida) for the US Dept. of Energy, Carbon Dioxide Research Division, Office of Energy Research, Washington DC.

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

Baker, J.T., Allen, L.H. Jr., Boote, K.J., Jones, J.W. and Jones, P. 1990a. Developmental responses of rice to photoperiod and carbon dioxide concentration. Agric. For. Meteorol. 50: 201-210.

Baker, J.T., Allen, L.H. Jr., Boote, K.J., Jones, P. and Jones, J.W. 1990b. Rice photosynthesis and evapotranspiration in subambient, ambient, and superambient carbon dioxide concentrations. Agron. J. 82: 834-840.

Baker, J.T., Allen, L.H. Jr. and Boote, K.J. 1992a. Response of rice to CO₂ and temperature. Agric. For. Meteorol. 60: 153-166.

Baker, J.T., Allen, L.H. Jr. and Boote, K.J. 1992b. Temperature effects on rice at elevated CO₂ concentration. J. Exp. Bot. 43: 959-964.

Baker, J.T., Laugel, R, Boote, K.J. and Allen, L.H. Jr. 1992c. Effects of daytime carbon dioxide concentration on dark respiration of rice. Plant Cell Environ. 15: 231-239.

Baker, J.T., Allen, L.H. Jr. and Boote, K.J. 1995. Potential climate change effects on rice: Carbon dioxide and temperature. In: Climate Change and Agriculture: Analysis of Potential International Impacts. C. Rosenzweig, L.A. Harper, S.E. Hollinger, J.W. Jones and L.H. Allen, Jr. (eds.). ASA Special Pub. No. 59, American Society of Agronomy, Madison, Wisconsin. pp. 31-47.

Baker, J.T., Allen, L.H. Jr., Boote, K.J. and Pickering, N.B. 1996. Assessment of rice responses to global climate change: CO₂. and temperature. In: Terrestrial Ecosystem Response to Elevated CO₂. G.W. Koch and H.A. Mooney (eds.). Physiological Ecology Series, Academic Press, San Diego. pp. 265-282.

Bazzaz, F.A., Jasienski, M., Thomas, S.C. and Wayne, P. 1995. Microevolutionary responses in experimental populations of plants in CO₂-enriched environments. Parallel results from two model systems. Proc. of the National Academy of Sciences S.92: 8161-8165.

Boote, K.J., Pickering, N.B., Baker, J.T. and Allen, L.H. Jr. 1994. Modelling leaf and canopy photosynthesis of rice in response to carbon dioxide. International Rice Research Notes 19: 47-48.

Broecker, W.S. 1987. Unpleasant surprises in the greenhouse? Nature 328: 123-126.

Bunce, J.A. 1990. Short- and long-term inhibition of respiratory carbon dioxide efflux by elevated carbon dioxide. Ann. Bot. 65: 637-642.

Campbell, W.J., Allen, L.H. Jr. and Bowes, G. 1988. Effects of CO₂ concentration on rubisco activity, amount, and photosynthesis in soybean leaves. Plant Physiol. 88: 1310-1316.

Campbell, W.J., Allen, L.H. Jr. and Bowes, G. 1990. Response of soybean canopy photosynthesis to CO₂, light, and temperature. J. Exp. Bot. 41: 427-433.

Crosson, P. 1993. Impacts of climate change on the agriculture and economy of the Missouri, Iowa, Nebraska, and Kansas (MINK) region. In: Agricultural Dimensions of Global Climate Change. H.M. Kaiser and T.E. Drennen (eds.). St. Lucie Press, Delray Beach, Florida. pp. 117-135.

Cure, J.D. and Acock, B. 1986. Crop responses to carbon dioxide doubling: A literature survey. Agric. For. Meteorol. 38: 127-145.

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

Curry, R.B., Peart, R.M., Jones, J.W., Boote, K.J. and Allen, L.H. Jr. 1990b. Response of crop yield to predicted changes in climate and atmospheric CO₂ using simulation. Trans. ASAE 33: 1383-1390.

Curry. R.B., Jones, J.W., Boote, K.J., Peart, R.M., Allen, L.H. Jr. and Pickering, N.B. 1995. Response of soybean to predicted climate change in the USA. 1995. In: Climate Change and Agriculture: Analysis of Potential International Impacts. C. Rosenzweig, L.A. Harper, S.E. Hollinger, J.W. Jones, and L.H. Allen, Jr. (eds.). ASA Special Pub. No. 59, American Society of Agronomy, Madison, Wisconsin. pp. 163-182.

Delucia, E.H., Sasek, T.W. and Strain, B.R. 1985. Photosynthesis inhibition after long-term exposure to elevated levels of atmospheric carbon dioxide. Photosyn. Res. 7: 175-184.

Dudek, D.J. 1989. Climate change impacts upon agriculture and resources: A case study of California. In: The Potential Effects of Global Climate Change on the United States. Appendix C, Agriculture, Vol. 1. EPA-230-05-89-053. J.B. Smith and D.A. Tirpak (eds.). US Environmental Protection Agency, Washington DC. pp. 5-1 to 5-38.

Hall, A.E. and Allen, L.H. Jr. 1993. Designing cultivars for the climatic conditions of the next century. In: International Crop Science I. D.R. Buxton, R. Shibles, R.A. Forsberg, B.L. Blad, K.H. Asay, G.M. Paulsen and R.F. Wilson (eds.). Crop Science Society of America, Madison, Wisconsin. pp. 291-297.

Hansen, J., Fung, I., Lacis, A., Lebedeff, S., Rind, D., Ruedy, R., Russell, G. and Stone, P. 1988. Global climate changes as forecast by the GISS 3-D model. J. Geophys. Res. 98: 9341-9364.

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

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

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

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

Jones, P., Jones, J.W. and Allen, L.H. Jr. 1985c. Carbon dioxide effects on photosynthesis and transpiration during vegetative growth in soybeans. Soil Crop Sci. Soc. Fla. Proc. 44: 129-134.

Keeling, C.D., Whorf, T.P, Wahlen, M. and van der Plicht, J. 1995. Interannual extremes in the rate of rise of atmospheric carbon dioxide since 1980. Nature 375: 660-670.

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., Mauney, J.R., Nakayama, F.S. and Idso, S.B. 1993. Effects of increasing atmospheric CO₂ on vegetation. Vegetatio 104/105: 65-75. Also: pp. 65-75. In: CO₂ and Biosphere. (Advances in Vegetation Science 14). J. Rozema, H. Lambers, S.C. van de Geijn and M.L. Cambridge (eds.). Kluwer Academic Publishers, Dordrecht.

Kramer, P.J. 1981. Carbon dioxide concentration, photosynthesis, and dry matter production. Bioscience 31: 29-33.

Long, S.P. 1991. Modification of the response of photosynthetic productivity to rising temperature by atmospheric CO₂ concentrations: Has its importance been underestimated? Plant Cell Environ. 14: 729-739.

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

Parry, M.L., Carter, T.R. and Konijn, N.T. (eds.). 1988. The Impact of Climatic Variations on Agriculture. Vols. 1 and 2. Kluwer Academic Publishers, Dordrecht.

Peart, R.M., Jones, J.W., Curry, R.B., Boote, K.J. and Allen, L.H. Jr. 1989. Impact of climate change on crop yield in the Southeastern U.S.A.: A simulation study. In: The Potential Effects of Global Climate Change on the United States. Appendix C, Agriculture, Vol. 1, EPA-230-05-89-053. J.B. Smith and D.A. Tirpak (eds.). US Environmental Protection Agency, Washington DC. pp. 2-1 to 2-54.

Pickering, N.B., Jones, J.W. and Boote, K.J. 1995. Adapting SOYGRO V5.42 for prediction under climate change conditions. In: Climate Change and Agriculture: Analysis of Potential International Impacts. C. Rosenzweig, L.A. Harper, S.E. Hollinger, J.W. Jones and L.H. Allen, Jr. (eds.). ASA Special Pub. No. 59, American Society of Agronomy, Madison, Wisconsin. pp. 77-98.

Poorter, H. 1993. Interspecific variation in the response of plants to an elevated ambient CO₂ concentration. Vegetatio 104/105: 77/97. Also: pp. 77-97. In: CO₂ and Biosphere. (Advances in Vegetation Science 14). J. Rozema, H. Lambers, S.C. van de Geijn and M.L. Cambridge (eds.). Kluwer Academic Publishers, Dordrecht.

Ritchie, J.T., Baer, B.D. and Chou, T.Y. 1989. Effect of global climate change on agriculture: Great Lakes Region. In: The Potential Effects of Global Climate Change on the United States. Appendix C, Agriculture, Vol. 1. EPA-230-05-89-053. J.B. Smith and D.A. Tirpak (eds.). US Environmental Protection Agency, Washington DC. pp. 1-1 to 1-42.

Rosenzweig, C. 1989. Potential effects of climate change on agricultural production in the Great Plains: A simulation study. In: The Potential Effects of Global Climate Change on the United States. Appendix C, Agriculture, Vol. 1. EPA-230-05-89-053. J.B. Smith and D.A. Tirpak (eds.). US Environmental Protection Agency, Washington DC. pp. 3-1 to 3-43.

Rosenzweig, C. and Iglesias, A. (eds.). 1994. Implications of Climate Change for International Agriculture: Crop Modelling Study. EPA 230-B-94-003. US Environmental Protection Agency, Washington DC.

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

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

Rosenzweig, C., Harper, L.A., Hollinger, S.E., Jones, J.W. and Allen, L.H. Jr. (eds.). 1995. Climate Change and Agriculture: Analysis of Potential International Impacts. ASA Special Pub. No. 59, American Society of Agronomy, Madison, Wisconsin.

Rowland-Bamford, A.J., Allen, L.H. Jr., Baker, J.T. and Bowes, G. 1991. Acclimation of rice to changing atmospheric carbon dioxide concentration. Plant Cell Environ. 14: 577-583.

Rozema, J., Lambers, H., van de Geijn, S.C. and Cambridge, M.L. (eds.). 1993. CO₂ and Biosphere. (Advances in Vegetation Science 14). Kluwer Academic Publishers, Dordrecht.

Satake, T. and Yoshida, S. 1978. High temperature induced sterility in indica rices at flowering. Japan. J. Crop Sci. 47: 6-17.

Smith, J.B. and Tirpak, D.A. (eds.). 1989. The Potential Effects of Global Climate Change on the United States. EPA-230-05-89-050. US Environmental Protection Agency, Washington DC.

Thorne, J.M. and Koller, R.M. 1974. Influence of assimilate demand on photosynthesis, diffusive resistance, translocation and carbohydrate levels of soybean leaves. Plant Physiol. 54: 201-207.

Valle, R., Mishoe, J.W, Campbell, W.J., Jones, J.W. and Allen, L.H. Jr. 1985. Photo-synthetic responses of 'Bragg' soybean leaves adapted to different CO₂ environments. Crop Sci. 25: 333-339.

Vergara, B.S. 1980. Rice plant growth and development. In: Rice Production and Utilization. B.S. Luh (ed.). Avi Publishing, Westport, CT. pp. 75-86.

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

Wittwer, S.H. 1995. Food, Climate, and Carbon Dioxide: The Global Environment and World Food Production. CRC Press/Lewis Publishers, Boca Raton, Florida.

Woodwell, G.M. and Mackenzie, F.T. (eds.). 1995. Biotic Feedbacks in the Global Climatic System: Will the Warming Increase the Warming? Oxford University Press, New York.

Wullschleger, S.D., Post, W.M. and King, A.W. 1994. On the potential for a CO₂ fertilization effect in forests: Estimates of the Biotic Growth Factor based on 58 controlled-exposure studies. In: Biotic Feedbacks in the Global Climatic System: Will the Warming Increase the Warming? G.M. Woodwell and F.T. Mackenzie (eds.). Oxford University Press, New York. pp. 85-107.

Yoshida, S. 1977. Rice. In: Ecophysiology of Tropical Crops. P.T. Alvim and T.T. Kozlowski (eds.). Academic Press, London. pp. 57-87.

Yoshida, S. 1981. Fundamentals of Rice Crop Science. International Rice Research Institute, Los Baños, Philippines.