Author for correspondence: Michael J. Ottman Tel: +520 621 1583 Fax: +520 6217186 Email:email@example.com
• Atmospheric CO2 concentration is expected to increase by 50% near the middle of this century. The effects the free air CO2 enrichment (FACE) is presented here on growth and development of field-grown grain sorghum (Sorghum bicolor) at ample (wet) and limiting (dry) levels of irrigation water at Maricopa, AZ, USA.
• Daytime CO2 mole fractions were 561 and 368 µmol mol−1 for the FACE and control treatments, respectively. Irrigation plus precipitation averaged 1132 mm for the wet plots and 396 mm in the dry plots.
• During the growing season, FACE increased biomass accumulation in the dry plots but the effects in the wet plots were inconsistent. At final harvest, FACE increased total yield from 999 to 1151 g m−2 in the dry plots and had no effect in the wet plots.
• If atmospheric CO2 continues to increase, total sorghum yield is likely to be higher in the future in areas where water is limited.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
Atmospheric CO2 concentration has increased dramatically since the Industrial Age and is projected to double by the end of this century (Houghton et al., 1990). Potential global warming and altered precipitation patterns due to increases in atmospheric CO2 concentrations are international concerns. However, potential positive benefits of increased CO2 are accelerated plant growth and improved agricultural productivity.
Doubling of CO2 concentration increases growth of C3 species by an average of 44% according to a review by Poorter et al. (1996). Species with a C4 photosynthetic pathway are expected to have a much smaller response to increases in CO2, and Poorter et al. (1996) have shown that the C4 growth response to increased CO2 is about half that of C3 plants. C3 species tend to dominate in temperate areas, whereas C4 species are more prevalent in tropical and subtropical areas. Over the past 125 years, C3 species have become more dominant compared with C4 species in the southern Great Plains and south-west USA. Mayeux et al. (1991) suggest that increasing CO2 is conferring a competitive advantage on C3 species in this situation.
The influence of elevated CO2 on sorghum grown with ample and deficit water is not clear based on previous work. Additional research is needed to provide more reliable data on the effect of elevated CO2 on sorghum growth and development and to improve growth models. The purpose of this research is to determine the effects of elevated CO2 under field conditions on sorghum biomass accumulation, grain yield, and phenological development with soil moisture as a variable.
Materials and Methods
Field studies were conducted to determine interactive effects of elevated CO2 and soil water on grain sorghum at the University of Arizona Maricopa Agricultural Center (MAC), Maricopa, Arizona, USA (Fig. 1). Experiments were conducted in 1998 and 1999 on a Trix clay loam soil (fine-loamy, mixed (calcareous), hyperthermic Typic Torrifluvents; Post et al. (1988)).
The free-air CO2 enrichment (FACE) technique was used to enrich the air in circular plots within a 12-ha sorghum field. Four replicate 25-m-dia. toroidal plenum rings constructed from 0.305-m inside diameter PVC pipe were placed in the field shortly after planting, and designated as F1 to F4 in Fig. 1. The centre-to-centre spacing of the rings was about 98 m. The rings had 2.5-m-high vertical stand pipes with individual valves spaced approx. every 2.4 m around the periphery (Wall & Kimball, 1993; Lewin et al., 1994). Air enriched with CO2 was blown into the rings, and it exited through tri-directional jets in the vertical pipes near the top of the crop canopy. CO2 concentration was measured in the centre of each array at 10 cm above the crop canopy. Wind direction and wind speed were measured just outside and to the north of each FACE ring. A computer-control system used the wind speed and CO2 concentration information to adjust the CO2 flow rates to maintain the desired CO2 concentrations at the centres of the FACE rings. The system used the wind direction information to turn on only those stand pipes upwind of the plots, so that CO2-enriched air flowed across the plots no matter which way the wind blew. When wind speeds were low (< 0.4 m s−1), and it was difficult to detect direction, the CO2-enriched air was released from every other vertical pipe around the rings. The CO2 flow rates were updated every second, and the choice of which vertical pipes to release from was updated every 4 s. With a similar apparatus and control strategy in cotton, we found that the 1-min-average CO2 concentrations were within 10% of the set point 87% of the time (Hendrey et al., 1993; Nagy et al., 1994).
Blowers were installed in the ambient control plots (marked C1 to C4 in Fig. 1) to provide air movement similar to that of the FACE plots. The control rings in the 1998 and 1999 FACE sorghum experiment had valves on all the vertical vent pipes, which were opened or closed according to the corresponding valve on the FACE ring in the same Rep. Use of these blowers was especially important at night in order to assure similar microclimatic conditions in both FACE and control plots (Pinter et al., 2000).
The FACE sorghum plots were enriched to a target of 200 µmol mol−1 CO2 above ambient. A separate sequential sampling system was used to measure the CO2 concentration in all of the FACE and control plots, as well as from two additional ambient sampling points (Fig. 1). Sixty seconds were required to measure the concentration in each plot, and 10 min for all 10 sites. The minimum value from among the most recent observations of the four control plots and the two ambient points was selected to provide ‘THE’ ambient value for the next 60 s against which to reference the 200 µmol mol−1 enrichment in the FACE plots. By selecting the minimum value, we generally chose the values from the most upwind plots, thereby avoiding contamination of the ambient value by CO2 from the FACE plots.
The FACE treatment was applied continuously from the date when 50% of the plants emerged until plant maturity. The daytime elevations of CO2 mole fraction in the FACE plots above the control plots were 193 µmol mol−1 (Table 1). These data also suggest the average contamination of the control plots with CO2 from the FACE plots was 7–8 µmol mol−1 above the upwind ambient concentration during daytime.
Table 1. CO2 concentrations, irrigation amounts, significant dates, fertilizer amounts, climatic conditions, and other cultural data for the 1998 and 1999 free-air CO2 enrichment (FACE) Sorghum experiments
a , at emergence; b , from the AZMET weather station located about 1 km from the FACE field (Brown, 1987). Growing degree days calculated with a base temperature of 7.0°C from emergence.
Each of the main circular FACE and control plots were split in to semicircular halves, with each half receiving either an ample (Wet) or a limited water (Dry) irrigation regime. The water was applied using flood irrigation. Only two irrigations were applied to the Dry treatments each season compared with seven (1998) or six (1999) to the Wet treatments. Irrigation water flow rates in the supply canal were determined from a gauge mounted in the wall of the canal upstream from a flume. After filling each carry ditch, irrigation amounts for a particular treatment strip were estimated using the time and flow rate relationship.
Irrigations for the Wet treatment were timed when 30% of the available water in the rooted zone was depleted. The plots were then irrigated with an amount calculated to replace 100% of the potential evapotranspiration since the last irrigation, adjusted for rainfall (Fox et al., 1992). Amounts were further adjusted to meet the minimum 100-mm irrigation required. The Dry plots were planned to receive one third the number of irrigations and water amount as the Wet plots, so as to severely drought the sorghum plants in the Dry plots, and such was achieved by applying water only twice, at the start and once again near mid-season.
The total amounts of irrigation plus rain applied during 1998 were 1218 and 474 mm to the Wet and Dry plots, respectively (Table 1). The 1218 mm applied to the Wet plots was probably in excess of evapotranspiration, and leaching below the root zone was estimated at 226 mm (Fox et al., 1992). In 1999, the amounts of irrigation plus rain were 1047 mm and 491 mm for Wet and Dry, respectively.
A sorghum crop had been grown on the land in the summer-fall of 1997, and a barley (Hordeum vulgare L.) crop had been grown in the winter-spring and harvested before maturity for hay at the beginning of April, 1998.
1998 crop culture
The field was disked in two directions and laser leveled before 8 April. Fertilizer (93 kg N ha−1 and 41 kg P ha−1 as urea (46–0-0) and monoammonium phosphate (11–52–0)) and pre-emergent herbicide (Dual 8E (metolachlor) at 2.7 kg ai ha−1)) were applied by air on 10 June and incorporated with a S-tine harrow (Model 20-ST, A. D. Williams, Casa Grande, AZ, USA) on 11 June. Certified grain sorghum seed (Dekalb DK54) was planted, which had been treated with fungicide (Apron, metalaxyl, Novartis, Basel, Switzerland) and safened (Concep III, fluxofenin, Novartis, Basel, Switzerland) for use with the herbicide. The seed was planted with a pneumatic planter on 15–16 July into dry soil in north–south rows spaced 0.76 m apart at a rate of 328 000 seeds ha−1 (10.9 kg ha−1; or 1 seed per 4 cm of row; Table 1). Erection of the FACE and control apparatus commenced immediately after planting and was completed by 27 July, when the first irrigation was applied to all plots. The FACE treatment started on 31 July, when slightly < 50% emergence had occurred. Stand counts shortly after emergence revealed a plant population of 223 100 plants ha−1.
As mentioned above, there was only one mid-season irrigation for the Dry plots, so a second application of fertilizer was applied to the Dry plots on 11 September 1998 (Table 1) at a rate of 186 kg N ha−1 to give a total N application of 279 kg N ha−1 for the season. The Wet plots were also irrigated and fertilized on this date, receiving 124 kg N ha−1. Because the Dry plots were more cracked, they were given more water for coverage, and they also received more N. To compensate, the Wet plots were given an additional 62 kg N ha−1 on their next scheduled irrigation day (day-of-year (DOY) 268) so that both Dry and Wet plots received that same total N for the season. The fertilizer was applied in irrigation water as Uran-32 [urea ammonium nitrate (32–0-0)].
FACE had little effect on heading, anthesis, and physiological maturity dates (Table 1). Several frosts occurred starting on 10 November until the end of the season that damaged leaves on the upper canopy. Final harvest was done on 21 December.
1999 crop culture
The sorghum stubble was chopped on 12 January, disked into the soil on 29 January, and disked a second time on 3 February. The field was re-leveled on 3 March. Fertilizer was applied by air on 1 June, again at a rate of 93 kg N ha−1 and 41 kg P ha−1. Herbicide (Dual, Novartis, Basel, Switzerland) was again applied and incorporated with an S-tine harrow (A. D. Williams). Planting was done on 14–15 June 1999, similarly to the prior year but a month earlier (Table 1). The seed was treated with fungicide (Apron, metalaxyl, Novartis) and safener (Concep III, fluxofenin, Novartis) as in 1998 but also with Gaucho (imidacloprid, Bayer Agricultural Products, Kansas City, MO, USA), an early season insecticide. The planting rate was 318 000 seeds ha−1 (9.97 kg ha−1; 1 seed every 4.1 cm of row), and the emerged population count was 259 500 plants ha−1. The first irrigation was applied on 28 June, and the 50% emergence date was 1 July. The FACE treatment commenced on 2 July. The mid-season irrigation for the Dry plots was on 6 August, and all plots were fertilized at that time with 172 kg N ha−1 in the irrigation water to give a total of 265 kg N ha−1 for the season.
A hailstorm damaged the upper leaves on 16 September, but the heads appeared unscathed. FACE had little effect on heading, anthesis, and physiological maturity dates (Table 1). Because of the earlier planting and the subsequently higher temperatures, the crop matured significantly earlier than in 1998. The final grain harvest was done on 26 October 1999.
Weekly plant sampling
Plants were sampled on a weekly basis from the 3- to 4-leaf stage until final harvest. A total of eight plants was sampled per semicircular plot. Four subplot sampling areas were designated within each plot that consisted of about a total of 7.2 m of row arranged in two or three rows. Two plants were removed each week from each of the four subplot sampling areas. Three plants were skipped between successive weekly samples. The plants were cut off at ground level and immediately placed in a cooled ice chest for transport from the field to the laboratory where they were stored at 5°C until processing. Plant height was measured from the base of the plant to the longest extended leaf or to the tip of the head, whichever was longer. The number of tillers and their developmental stages were recorded. Leaves were removed from the stem at the collar and separated into green and brown leaves. Leaf area of green leaves was measured with a leaf area meter (Li-Cor 3100, Li-Cor Inc., Lincoln, NE, USA). Heads were removed from the stems. Stem area was estimated from stem length (without the head) × diameter. The stems, brown and green leaves, and heads were dried separately in an oven at c. 65°C for 2–4 d and weighed. The masses of these various plant parts were adjusted to a ground area basis using plant density determined at harvest. Leaf area index and stem area index were also calculated using plant density at harvest. Specific leaf area was calculated by dividing green leaf area by green leaf weight.
Phenology was determined according to Vanderlip (1993). Before heading, developmental stages were based on the appearance of the leaf collars on eight plants adjacent to the sampling areas. After heading, phenological stages were determined on plants sampled for biomass. Numbers ranging from 1 (50% of plants heading) to 7 (50% of plants at physiological maturity) were assigned to designate growth stage after the vegetative period. Before heading, growth stages represent mean leaf number of all plants and not the most advanced 50% as was done after heading.
Grain yield was determined from a final harvest, nontraffic area composed of six adjacent rows, each 5 m in length. Heads were removed from the plants, weighed, and threshed using a small plot harvester. The grain was weighed and a sample of grain and chaff (nongrain portion of the head) were saved for moisture determination. Grain moisture was determined with a digital moisture meter and chaff moisture was determined from wet and oven dry (65°C) weights. Headless plants from the centre two rows of the nontraffic area were cut at ground level and harvested for end of season biomass yield. Moisture was determined from wet and oven dry (65°C) weights. Stover yield was defined as nongrain yield and was calculated as the sum of the head-less-plant biomass plus chaff yield. Total aboveground biomass (i.e. total yield) was summed from grain yield plus stover yield. Harvest index was calculated by dividing grain yield by total aboveground biomass. The number of kernels in a 10-g sample was counted and the weight per kernel determined. The number of heads in the final harvest area was counted. Kernels-per-head was calculated by dividing grain weight per head by kernel weight. Kernels-per-unit-area was calculated by multiplying kernels-per-head by heads-per-unit area.
Statistical analysis methodology
The data were analysed as a strip-split-plot design using the ‘Mixed’ Procedure (Littell et al., 1996) with CO2 as the main plot and irrigation as the strip-split. The significance level chosen was P = 0.10. Weekly plant growth is presented as a 3-wk running average, but the statistics were run on the original data using repeated measures analysis and specifying first order, autoregressive correlation for the covariance structure.
Final yield and grain yield components
FACE increased stover yield in both years, but the effects on grain and total yields were dependent on year and water regime (Table 2). Grain, stover, and total yield increases due to FACE occurred more often in the Dry than the Wet plots. Averaged over years and water levels, stover yield was increased by 9% by FACE, while grain and total yield were not affected by FACE. In 1998, FACE treatment increased stover and total yield but did not affect grain yield and harvest index. In 1999, FACE treatment increased stover yield, but not grain yield, total yield or harvest index. Carbon dioxide–water interactions were detected in 1999 for grain yield and harvest index. In 1999 the FACE effect was not significant for grain yield in the Dry treatment but in the Wet treatment grain yield was actually decreased by 11% by FACE. Harvest index averaged across years was not affected by FACE in the Dry treatment but decreased in the Wet treatment similar to grain yield. Even though the FACE effects were not significant for grain yield within each year, the ANOVA combined over years resulted in a significant CO2–year interaction. FACE resulted in an 8% grain yield increase in 1998, but a 3% grain yield decrease in 1999 averaged over water levels. A three–way interaction among CO2, water, and year resulted for total yield since FACE increased total yield 7–18% except for the Wet plots in 1999 where total yield decreased 1%.
Table 2. Sorghum yield at final harvest as affected by water and CO2. Values in parentheses represent percentage change due to free-air CO2 enrichment (FACE)
Grain (g m−2)
Stover (g m−2)
Total (g m−2)
ns, + , *, ** = not significant at P = 0.10 and at P < 0.10, 0.05, and 0.01, respectively.
Grain yield differences in our study can be explained primarily by kernel weight, kernels per head, and kernels per m2, but not by heads per m2 (Table 3). In 1998, FACE increased the number of kernels per head and kernels per m2 in the Dry treatment, but had no effect in the Wet treatment, reflective of grain yield. Kernel weight was not affected by FACE in 1998. In 1999, FACE increased kernel weight in the Dry treatment, but decreased kernel weight in the Wet treatment similar to grain yield trends. Carbon dioxide had no effect on the number of heads per m2 in either year. Water had a much greater influence on grain yield than CO2, and weight per kernel and kernels per head accounted for about two-thirds and one-third of the grain yield increase in 1998, and vice-versa in 1999.
Table 3. Sorghum yield components at final harvest as affected by water and CO2. Values in parentheses represent percentage change due to free-air CO2 enrichment (FACE)
Kernel weight (mg)
Kernel number (head−1)
Kernel density (m−2)
Head density m−2
ns, + , *, ** = not significant at P = 0.10 and significant at P < 0.10, 0.05, and 0.01, respectively.
Water regime had a much greater influence on dry matter accumulation than FACE, but FACE resulted in some notable effects. FACE increased the mass of various plant parts in the Wet and Dry plots and decreased mass in the Wet plots in several instances in 1999. The weight of green leaves was increased by FACE at several sampling times during 1998 (Fig. 2a), but in 1999 FACE decreased green leaf weight in the Wet plots after anthesis (Fig. 3a). FACE increased the amount of brown leaves on the plant (Figs 2b; 3b), especially for the Wet plots in 1999. FACE increased stem weight at various times during the season in both Wet and Dry plots (Figs 2c; 3c). Head weight was not affected by FACE in 1998. In 1999, head weight was decreased by FACE in the Wet plots and, at one stage, head weight was increased by FACE in the Dry plots (Figs 2d; 3d). Total biomass was increased by FACE in the Wet and Dry plots at a few sampling times (Figs 2e; 3e). Total biomass appears to decrease due to FACE in the Wet plots after anthesis in 1999 but the results were significant only at a single sampling time.
Leaf, stem, and plant development
FACE had relatively minor effects on leaf, stem, and plant development compared with water, but the FACE effects are important, nevertheless. FACE had a stimulatory effect on leaf, stem, and plant development in general. FACE increased LAI except in 1999 on the Wet plots where LAI was decreased by FACE (Figs 4a; 5a). Specific leaf area was increased by FACE in the Dry plots both years (Figs 4b; 5b). Stem area index was increased by FACE especially in 1998 (Figs 4c; 5c). FACE increased stems per plant especially before anthesis (Figs 4d; 5d). Stem height was fairly consistently increased by FACE (Figs 4e; 5e). FACE delayed leaf appearance in the Wet plots in 1998, accelerated leaf appearance in the Wet plots in 1999, and delayed early grain development and accelerated later grain development in the Dry plots in 1999 (Figs 6; 7).
Final yield and grain yield components
Idso & Idso (1994) indicate that growth enhancement from increased CO2 is greatest when water is most limiting. This was indeed the case in our study where FACE increased total yield by 15% in the Dry plots, averaged over years, and had no effect in the Wet plots. Other studies have shown sorghum biomass was not affected by differences in CO2 with ample water (Mauney et al., 1978; Marc & Gifford, 1984; Ellis et al., 1995). Under drought, however, a doubling of CO2 concentration resulted in a 26% increase in total yield in the study of Morison & Gifford (1984b) and a 31% increase in grain yield and 34% increase in total yield in the study of Chaudhuri et al. (1986). In our study averaged over years, FACE decreased harvest index in the Wet plots but did not affect harvest index in the Dry plots. The decrease in harvest index in the FACE Wet plots coincided with reduced kernel weight in these plots, particularly in 1999. The decrease in kernel weight in the Wet plots in 1999 may have been related to premature leaf senescence and increased necrotic leaves during the grain-fill period, as discussed in the next section.
Dry matter accumulation
The increases in total biomass due to FACE measured later in the season in the Dry plots but not the Wet plots were consistent with final yields and expected based on previous reports in the literature (i.e. Idso & Idso, 1994). The differences in total biomass were closely related to stem biomass, a large component of total biomass. FACE decreased head and green leaf yield in 1999 in the Wet plots. Ellis et al. (1995) also reported a negative effect of CO2 on sorghum growth with ample water. FACE reduced stomatal conductance in our study (Wall et al., 2001) and may have had a negative impact on physiological processes in Wet plots. In the Dry plots, reduced stomatal conductance due to FACE probably contributed to higher soil water levels measured later in the season (Conley et al., 2001) and the yield increases measured near the end of grain fill.
Leaf, stem, and plant development
Leaf and stem areas were affected by FACE similar to biomass of these plant parts. FACE delayed loss of green leaf area in the Dry plots and accelerated loss of green leaf area in 1999 in the Wet plots, similar to our findings for green and brown leaf biomass. The loss of LAI in the FACE Wet plots at the end of the season in 1999 may explain the reduced kernel weight obtained that year. FACE increased specific leaf area in the Dry plots. This may be related to improved water relations due to conservation of soil moisture from partial stomatal closure. FACE increased stem area, particularly later in the season in 1998, suggesting the FACE effect may be cumulative.
Each plant produced about two tillers that eventually withered away by the end of the season. FACE generally increased stems per plant early in the season especially in the Dry plots.
FACE did not have a consistent effect on sorghum phenological development in our study. Sorghum development should be related to the temperature of the developing meristem. However, the elevated temperatures due to drought or stomatal closure may exceed an optimum threshold so that there is no net effect on development. Alternatively, night temperatures, independently of stomatal responses, may be more critical for development.
By the middle of this century when atmospheric CO2 reaches the levels simulated by the FACE treatment in this study, total sorghum yields are likely to be higher than present in areas where sorghum is grown with limited water. In high rainfall or irrigated areas, total sorghum yields may not be affected by increased atmospheric CO2.
The research was supported by Interagency Agreement No. DE-AI03–97ER62461 between the Department of Energy, Office of Biological and Environmental Research, Environmental Sciences Division and the USDA, Agricultural Research Service (Bruce A. Kimball, PI); by Grant no. 97–35109–5065 from the USDA, Competitive Grants Program to the University of Arizona (Steven W. Leavitt, PI); and by the USDA, Agricultural Research Service. It is part of the DOE/NSF/NASA/USDA/EPA Joint Program on Terrestrial Ecology and Global Change (TECO III). This work contributes to the Global Change Terrestrial Ecosystem (GCTE) Core Research Programme, which is part of the International Geosphere-Biosphere Programme (IGBP). We also acknowledge the helpful cooperation of Dr Robert Roth and his staff at the Maricopa Agricultural Center. Portions of the FACE apparatus were furnished by Brookhaven National Laboratory, and we are grateful to Mr Keith Lewin, Dr John Nagy, and Dr George Hendrey for assisting in its installation and consulting about its use. The technical assistance of Mark Rogers, Carrie O’Brien, Steve Waichulaitis, and Chandra Holifield is greatly appreciated.