Content uploaded by Hugo Rogers
Author content
All content in this area was uploaded by Hugo Rogers on Nov 18, 2017
Content may be subject to copyright.
Agriculture, Ecosystems and Environment, 16 (1986) 113--128
Elsevier Science Publishers B.V., Amsterdam -- Printed in The Netherlands 113
SOYBEAN GROWTH AND YIELD RESPONSE TO ELEVATED CARBON
DIOXIDE
HUGO H. ROGERS ~
USDA-ARS, National Soil Dynamics Laboratory, Auburn, AL 36831-0792 (U.S.A.)
JENNIFER D. CURE
Botany Department, Duke University, Durham, NC 27706 (U.S.A.)
JOY M. SMITH
Department of Soil Science, North Carolina State University, Raleigh, NC 27750 (U.S.A.)
(Accepted for publication 27 January 1986)
ABSTRACT
Rogers, H.H., Cure, J.D. and Smith, J.M., 1986. Soybean growth and yield response to
elevated carbon dioxide.
Agric. Ecosystems Environ.,
16: 113--128.
Soybeans
(Glycine
max L. Merr. 'Bragg') were grown in seeded rows in open-top field
chambers and exposed continuously to a range of elevated CO2 concentrations through-
out the 1982 and 1983 growing seasons. During 1983, a water stress treatment was also
imposed.
Comparison of vegetative growth with a similarly conducted pot experiment showed
an increased ratio of leaf area to total top dry weight in the seeded row plants, but
generally similar qualitative effects of elevated COs. Careful recording of mainstem leaf
emergence rates and reproduction stages showed no consistent effect of CO 2 under well
watered conditions, but in 1983 there was a distinct modification by high CO~ of the
water stress-induced hastening of the time to physiological maturity.
In 1982, and for the well watered plants in 1983, standing biomass at maturity was
increased significantly by elevated CO2, but harvest index decreased and yield was (sta-
tistically) unaffected by the treatment. The yield responses calculated for a doubling
of the current COs concentration for these well watered treatments were 1.07 and 0.93,
respectively. In the water stress treatment in 1983, however, harvest index did not
decrease in the presence of elevated CO2, and a highly significant yield response occurred
(1.41 at 700 #l 1-').
INTRODUCTION
Recent global carbon cycle models project a range of possible atmos-
pheric carbon dioxide (CO2) concentrations in year 2075 of approximately
500--1500 pl 1 -~, with a median of about 700 pl 1-1 (Edwards et al., 1984).
In order to evaluate the impact of this change on agricultural productivity,
'Formerly USDA-ARS, Raleigh, NC, U.S.A.
0167-8809/86/$03.50 © 1986 Elsevier Science Publishers B.V.
114
work has recently focused on the direct effects of elevated CO2 on long
term growth and yield of crop plants (Lemon, 1983). Until very recently,
however, field studies have been lacking due to a technical inability to
generate the large scale test atmospheres required. In recent years this has
been overcome at several research sites (Kimball, 1983; Rogers et al., 1983b;
Havelka et al., 1984) and equipment for the study of crop response to
CO2 enrichment in the field is now available.
Soybeans were exposed to a range of above-ambient concentrations of
CO2 in open-top field chambers throughout the 1982 and 1983 growing
seasons, and in 1983, a water stress treatment was also included. In this
paper, vegetative growth observations are compared with those from a
previous COs enrichment study for soybeans conducted in pots in open-top
chambers, and the yield data from the field are presented against a back-
ground of yield results obtained under a wide variety of environmental
conditions.
MATERIALS AND METHODS
Soybean plants
(Glycine
max L. Merr. 'Bragg') were grown in open-top
field chambers (Rogers et al., 1983b). Each chamber consisted of a cy-
lindrical aluminum frame (3 m in diameter X 2.4 m in height) covered
with PVC film Roll-A~Glass with a 45 ° frustum attached at the top. Air
with the desired CO2 concentration was supplied day and night through
perforations in the inner wall of the lower half of the chamber. Air was
adjusted to the proper CO2 level with pure CO2 fed from a supply tank
of pure liquid CO2. Mixing occurred in a fan-driven plenum box where
the air and CO2 were brought together and blown into the chamber (Rogers
et al., 1983b). Gas samples were drawn from each plot at 10 cm above
canopy level 3 times hourly and adjustments in CO2 dispensed to each
chamber were made twice daily based on the most recent 3-h means. A full
account of CO2 measurement and control has been given (Rogers et al.,
1983b).
1982 Experiment
In 1982, there were two replicate plots per CO2 treatment, randomly
arranged in each of two blocks. Seasonal daytime 0500--1900 h EST)
mean CO2 concentrations for the 6 CO2 treatments were 348 pl 1-1 (open
plot without chamber), 349 ul 1-1 (ambient chamber), 421 ~11-1, 486 pl 1-1,
645 ~1 1-1 and 946 ~1 1-1. Extremely wet weather delayed planting until
29 June. Six days after planting (DAP), the chambers were set in place
and at 8 DAP, CO2 dispensing and monitoring began. Water was applied
to the plots whenever tensiometers (one at 30 and one at 45 cm) showed
soil moisture tensions greater than 50--60 centibars. From 1 of the 2 rep-
licate plots within each block, sequential harvests were made for growth
analysis. These harvests were of at least 8 plants each (the first 2 were
115
thinning harvests) and occurred on Days 14 (seedling), 29 (mid-vegetative),
76 (early pod fill), 125 (physiological maturity) and 140 (harvest maturity,
when all pods were brown). The other replicate plot within each block
was reserved for yield harvests only (the last 2 dates). The yield harvests
at 125 and 140 DAP consisted of at least 0.75 m of row 0.3 m inside the
chamber wall from each of the 2 replicates within each block. Seeds were
removed from pods harvested at 140 DAP for calculation of harvest index
(seed dry wt/total top dry wt).
1983 Experiment
In 1983 there were two blocks, with one replicate of each of two watering
regimes randomly arranged in each block. The seasonal daytime 0500--
1900 h EST) means for the 5 CO2 treatments were 349 ~1 1-1 (open plot
without chamber), 346 pl 1-1, 424 pl 1-1, 505 gl 1-1 and 650 ~11-1. Planting
occurred 6 June, and chambers were in place and CO: dispensing and moni-
toring began by 10 DAP. Semi-open platforms were placed between the
rows to minimize compaction of soil between the rows. Rain covers were
placed over the tops of the stressed plots during rain, or overnight if rain
threatened, in such a way that the air flow through the chambers was not
affected. Non-stress plots were irrigated whenever tensiometers (2 at 30 cm
and 2 at 45 cm depth in each plot) reached 20--30 centibars. For the first
50 days, only a mild water stress was permitted to develop in the stress
plots, irrigation taking place when the tensiometers showed soil moisture
tensions of 70--80 centibars. After 50 DAP irrigation took place only on
days when plants were seen to be wilted in the early mornings.
In both years, plants were sprayed weekly with appropriate insecticides
and weeds were controlled by hand within the test plots. Plants were tied
up to avoid lodging. Stem and leaf samples were oven-dried for at least 72 h
at 55 + 5°C and pods were dried at about 21°(:;. Leaf areas were measured
photometrically with automatic area meters. In both years, plants were
thinned to a density of 15 m -1, and rows were 96.6 cm apart. However,
in 1982, the chambers were placed over the rows such that two main rows
of maximum length were centered in the chamber; growth of the two
"border rows" was necessarily disrupted by chamber walls and consequently
their usefulness as border rows was compromised. In 1983, the chambers
were placed so that there was one main row down the center of the chamber
with two good border rows. Samples were collected only from the center
rOW.
Growth Analysis
The following growth functions were calculated from above-ground mass
data according to Kvet et al. (1971):
NAR (mean net assimilation rate) = dry matter accumulation rate per unit
leaf area
116
LAR (mean leaf area ratio) = ratio of leaf area to total top dry matter
RGR (mean relative growth rate) = dry matter accumulation rate per unit
dry matter
NAR, LAR and RGR from a 1981 pot experiment (see Rogers et al., 1984a)
were recalculated deleting root dry weights for comparison with values
obtained from the 1982 field plots. (For detailed data from 1982 field study
see Rogers and Bingham, 1982). In the pot study, the intervals were (1)
5--14 days, (2) 14--49 days and (3) 49--84 days. In the field study the
intervals were (1) 5--14 days, (2) 14--29 days and (3) 29--76 days. Thus,
although in both studies the same genotype and the same exposure system
were used, they were performed in different years. Also, the growth char-
acteristics of seeded row plants in the second and third intervals reflect the
behavior of younger plants than those in the pots, as well as the behavior
of plants grown without apparent restriction of root growth.
Statistics and Response Ratio Calculations
Regression analyses were performed by the least squares method (Neter
and Wasserman, 1974). In Tables, I, II and IIIa significant water stress
TABLE I
1982 harvest data for 'Bragg' soybeans grown in open top chambers at 5 CO: concentra-
tions. N = 8
CO: Stem dry wt. Pod dry wt. Pod number Harvest
(pl 1-1) 1 (g m -~) (gm -1) (m -1) index 2
348 263 438 1032 0.47
349 395 744 1509 0.49
421 456 803 1698 0.49
496 482 739 1658 0.45
645 526 835 1854 0.46
946 636 873 2173 0.42
s~ 25 57 107 0.01
CV (%) 13 14 15 4.3
b o 132_+ 31 438- + 57 666-+ 126 0.51- + 0.01
bcharnbez 152+303 360±633 512-+1213 0.01±0.01
blmeaz4 377-+52 NS 1049±214 -0.01-+0.02
bquadrati c NS NS NS SS
R ~ 0.99 0.94 0.99 0.88
1The first CO2 value (348) is from the open plots (no chambers); other values are from
within chambers. Values for CO2 are seasonal daytime means.
2Harvest index is from sampling at harvest maturity only. Other variables represent
average values from harvests at physiological maturity and harvest maturity.
3Significant F (0.95 level) for chamber effects.
*Linear coefficients and their standard errors should be multiplied by 10-3.
TABLE II
1983 harvest data for stressed (S) and non-stressed (NS) "Bragg" soybeans grown in open plots and in open-top chambers at 4 CO 2 concentrations
N=2
CO 2 Stem dry wt. Pod dry wt. Pod number Harvest
(~ul I-I) I (g m -I) (g m -I) (m -]) index
S NS S NS S NS S NS
349 317 425 335 405 759 1004 0.36 0.35
346 339 488 464 683 991 1506 0.40 0.42
424 432 613 584 699 1049 1509 0.40 0.38
505 426 603 602 661 1120 1459 0.41 0.38
650 522 682 661 654 1325 1510 0.40 0.35
s~ 32.8 28.5 60.'/ 0.01
CV (%) 9.6 7.0 '7.0 3.9
b o 101-+52 225-+52 131-+52 405+28 516-+81 864-+81 0.40+0.01
bchambe r 69-+352 165-+36 = 269-+322 354-+ 542 0.05-+0.012
blinear s 553+123 585+127 NS 550+190 --0.11-+0. °3
b quadratic NS NS NS NS NS
R 2 0.96 0.98 0.99 0.97 0.90
0.38+0.1
IThe first C02 value (349) is from the open plots (no chambers); other values are from within chambers. Values for CO= are seasonal daytime means.
2Signlficant F (0.95 level) for chamber effect.
3Linear coefficients and their standard errors should be multiplied by 10-3.
b-a
b~
TABLE Ill
Protein, off and fiber content of 'Bragg' soybean seeds produced in open top chambers at v~ious CO 2 concentrations in 1982
(N=8) and 1983 (Nffi2). In 1983 S ffi water stressed and NS = non-stressed treatments I
1982 1983
CO 2 Protein Oil Fiber CO 2 Protein (%) Oil (%) Fiber (%)
(~ I -I) (%) (%) (%) (~{~ I -I)
S NS S NS S NS
348 40.6 17.6 41.8 349 44.8 43.4 18.0 18.9 37.1 38.0
349 40.8 17.5 41.7 346 41.6 43.0 19.4 19.0 39.1 38.0
421 40.0 17.6 41.4 424 42.8 43.1 19.0 18.9 38.3 38.0
496 41.2 17.2 41.6 505 41.5 43.0 19.7 18.6 39.0 38.4
645 41.1 17.3 41.6 650 41.1 42.5 19.4 19.1 40.0 38.4
946 41.2 17.0 41.9
s~ 0.25 0.24 0.12 0.39 0.37 0.39
CV (%) 1.2 3.4 0.9 1.29 2,72 1.44
b o 40.8 17.4 41.7 43.0 44.8-+0.4 19.0 38.7 37.1-+0.4
bchambe r NS NS NS NS -3.1-+0.4 NS NS 1.8-+0.43
bb~tnea r NS NS NS NS NS NS NS NS
quadratic NS NS NS NS NS NS NS NS
R 2 NS NS NS NS 0.86 NS NS 0.99
1Analysis performed by Dr. Jim Cavins, Research Chemist, Horticultural and Special Crops Laboratory, Peoria, IL 61604.
2The first CO 2 value is from the open plots (no chambers); other values are from within chambers. Values for CO~ are seasonal
daytime means.
3Significant F (0.95 level) for chamber effect.
TABLE IV
Yield Response Ratios (1000/~l 1-'/350 ~l 1-1) and changes in harvest index for soybeans tested at COs concentrations > 1000 ~11-1
under different experimental conditions
Experiment Genotype I Growth condition CO 2 treatments Response Change in HI
(#11-~) 2 ratio
1000/350
la Maddox, 1974 Bragg pot with vermiculite and perlite; 360, 1300 1.13 0.59-*0.58
(VII, d) small chambers within a greenhouse,
enriched after 12 days post-flowering
lb Maddox, 1974 Forrest same conditions as above 360, 1300 1.02 0.58-*0.59
(V,d)
lc Maddox, 1974 D69--B5 same conditions as above 360, 1300 1.05 0.60--0.61
(VI, d)
ld Maddox, 1974 Lee 68 same conditions as above 360, 1300 0.97 0.59-*0.60
(VI, d)
2a Cooper and Brun, 1967 Hark 3 plants per pot with soil; supple- 350, 1350 1.37 0.61~0.57
(I, i) mented winter greenhouse light
2b Cooper and Brun, 1967 Chippewa same conditions as above 350, 1350 1.26 0.59-*0.55
64 (I, d)
3 I-Iardman and Brun, 1971 Hark spaced plants in ground in open-top 350, 1200 1.28 0.32-*0.27
(I, i) chambers
4 Ackerson et al., 1984 Wye seeded rows, open-top chambers "AMB" (-- 350), 1.20 0.42-~0.43
(IV, i) 1200
5 Havelka and Hardy, 1976 Kent & seeded rows, open-top chambers "AMB" (= 350), 1.78 0.4740.56
Clark (IV, (enriched Days 35--105) "800--1200" (pods) (pods)
d) (= 1000)
6a Havelka et al., 1984 Kent seeded rows, open-top chambers 330, 1200 1.60 0.32-*0.34
(IV, d)
(enriched Days 22--maturity)
6b Havelka et al., 1984 Ware same conditions as above 330, 1200 1.41 0.34,0.39
(IV, d)
1The Roman numeral refers to maturity grouping and 'd' or 'i' refers to determinate or indeterminate growth habit, respectively.
2From 350 to 1000 ppm CO2.
Q
TABLE V
Yield Response Ratios (700 ul 1-1/350 ~1 1-1) and changes in harvest index for soybeans tested at CO2 concentrations < 1000 ul 1-1
under different experimental conditions
Experiment Genotype Growth condition CO2 treatments Response Change in HI
(~l l- 1) ratio
700]350
la Sionit, 1983 Ransom pot with gravel and turface; 350,675 2.34
(VII, d) 1 controlled environment; light
550 uE m -~ s-l; 1/2 Hoagland's
same conditions as above except 350, 675
1/8 Hoagland's
SPAR units, seeded rows
lb Sionit, 1983 Ransom
(VII, d)
2 Jones et al., 1984 Bragg 330,
(VII, d) 600,
3 Acock et al., 1982 Forrest SPAR units, seeded rows 330
(V, d) 600
4 Acock et al., 1983 Bragg SPAR units, seeded rows 330.
(VII, d) 600
5 Rogers et al., 1983a Ransom pots with soil in open-top chambers 340
(vii, d) 718
6 Rogers et al., 1981 Bragg pots with soil in open-top chambers 332
(VII, d) 623
0.73-*0.69
2.09 0.73~0.73
450, 1.35 0.31~0.30
800
450, 1.38 0.44-+0.45
8OO
450, 0.14 0.79-*0.73
800 (pods) (pods)
520, 1.31 0.58-*0.55
910
428, 534, 1.16 0.51--,0.47
772, 910
7a Rogers and Bingham, Bragg
1982 (VII, d)
7b Rogers and Bingham, Bragg
1982 (VII, d)
8a Israel and Rogers, Bragg
1982 (VII,
d)
8b Israel and Rogers, Bragg
1982 (VII,
d)
9 Rogers, Cure and Smith Bragg
~this report) (VII, d)
10a Rogers, Cure and Smith Bragg
(this report) (VII, d)
10b Rogers, Cure and Smith Bragg
(this report) (VII, d)
pots with soil in open-top chambers; 349, 421,496, 1.03 0.50-*.0.43
well watered 645, 946
same condition as above except 349, 421, 496, 1.14 0.54-*0.47
water stressed 645, 946
pots with perlite; open-top chambers; 349, 421, 496, 1.08 -- --
Rhizobium
strain USDA 110 645, 946
same conditions as above except 349, 421, 496, 0.95 -- --
Rhizobium
strain USDA 31 645, 946
seeded rows, open-top chambers 349, 421,496, 1.07 0.49-*0.45
(1982) 645, 946
seeded rows, open-top chambers 346, 424, 0.93 0.41~0.34'
(1983); well watered 505, 650
same as above except water stressed 346, 424, 1.43 0.40-*0.40
505, 650
1The Roman numeral refers to maturity grouping, and 'd' or 'i' refers to determinate or indeterminate growth habit, respectively.
From 350 to 700 ppm CO2.
b.a
tO
k.A
122
effect was indicated for a variable by the presence of separate Y-intercepts
(bo) for the stressed (S) and non-stressed (NS) treatments, and a CO2 X
stress interaction was also described by separate parameter estimates for S
and NS plants.
In Tables IV and V, seed yield data from the listed references were re-
gressed against CO2 concentration, and yield response ratios were calculated
from predicted yield values for 1000 and 700 pl 1-1 CO2, respectively, relative
to yield at 350 pl 1-1 CO:. Linear regressions were employed where only
two data points were provided (Table IV). Either linear or quadratic models
were used, where statistically appropriate, in cases where more data points
permitted (Table V). Since yield response to CO2 concentration departs
from linearity in the range 350--1300pl 1-1, the response ratios at 1000~11-1
(based on data obtained at 350 and 1300~11-1) are probably underestimates.
Nevertheless, the ratios in Table V are internally comparable.
RESULTS AND DISCUSSION
Row Crop Studies: 1982--1983
In 1982, stem dry weight and pod number increased in a linear fashion
with increasing CO2 concentration as denoted by significant, positive linear
regression coefficients in Table I. Harvest index (HI) decreased, however,
and although there was a definite trend towards increasing pod dry weight,
no statistically significant effect of CO2 enrichment on yield was observed.
A similar pattern of responses occurred for the non-stressed plants in 1983,
i.e. an increase in stem dry weight with increasing CO2 concentration,
together with a decreasing harvest index and a lack of yield response to
CO2 (Table II). However, the stressed plots in 1983 showed clear effects of
the chronic water stress on stem dry weight and pod number, as denoted
by the presence of different intercepts (bo) , but no stress × CO2 interaction.
For pod yield, however, our analysis indicated both a water stress effect
and a significant CO2 × water stress interaction, resulting in separate equat-
ions for the two treatments. At 350 pl 1-1, the pod dry weight decrease due
to water stress was about 175 g m -~, but this effect became insignificant
at higher CO2 concentrations.
Seed protein and oil content were lower in 1982 than in 1983, and there
was a significant effect of water stress on protein content in 1983, but
there was no effect of elevated CO2 on seed composition in either year
(Table III). Moreover, further analysis of the oil fraction of the 1982 seed
showed no effect of elevated CO2 on fatty acid composition (R. Wilson,
unpublished data, 1983).
Using predicted values obtained from the parameter estimates in Tables
I and II, total standing biomass produced in ambient-level CO2 chambers
was quite similar between the 1982 crop and the non-stressed 1983 crop
(1189 g m -~ in NS 1983 plots vs. 1139 in 1982). However, HI was much
123
greater in 1982 and yield was therefore greater. The reason for this dif-
ference in HI is uncertain, but it may be related to the different row con-
figurations within the chambers described in "Methods".
Figure 1 illustrates the relative effects of continuous exposure to elevated
CO2 on the yield of soybeans grown under conditions as close as possible
to field conditions. Water stress was not imposed in 1982. Only in stressed
plots in 1983 was CO2 effective in increasing pod yield, suggesting that
increased yield brought about by elevated CO2 is largely due to changes in
plant water relations.
1.4
1.3
e-
1.2
1.1 P.*
"0
.~
>. 1.0
0.9
0.8
i i | ~* •
+ + 75 + 150 + 300 + 600
CO 2 Treatment ( #l
1-1
above ambient level)
Fig. 1. Relative pod yield for 'Bragg' soybeans grown
in the
field in open-top chambers
at elevated CO2. The 1982 experiment did not include water stress treatments.
Elevated CO2 had a small accelerating effect on the rate of leaf initiation
in both years, resulting in the addition of one vegetative mainstem node
before meristems were converted to the reproductive mode (21 vs. 20
nodes in 1982; 23 vs. 22 nodes for non-stressed plants in 1983). There
was also a trend toward slightly faster expansion of the leaves in high CO2
in both years. Although the same number of mainstem leaves was eventually
present in the stressed vs. non-stressed plants in 1983, the stress treatment
slowed production of the last leaves by almost a week at low CO2 and by
3 days at 650 ppm. All reproductive stages in high CO2 occurred slightly
behind those for control plants in 1982. In 1983, however, physiological
maturity was accelerated by 4 days by high CO2. Water stress also acceler-
ated maturity, by 7 days at the low CO2 concentration, but this effect of
water stress on time to maturity was not observed at high CO2. Yellowing
124
of leaves was observed to occur more rapidly at high CO2 concentrations
both years.
Comparison with Pot Experiments
Since so much CO2 work has been done with potted plants, it was of
interest to compare some vegetative growth characteristics of 'Bragg' soy-
beans obtained in the 1982 field study with those obtained with the same
genotype similarly exposed to CO2-enriched air in open-top field cham-
bers, but grown in pots. The objective was to compare the effect of
elevated CO2 on RGR (mean relative growth rate), NAR (mean net as-
similation rate) and LAR (mean leaf area ratio) of plants grown in 10-inch
pots (see Rogers et al., 1984a) with plants grown with no apparent root
restriction. These vegetative growth functions were calculated according to
Kvet et al. (1971) over three intervals (see "Methods").
1st Interval 2nd Interval 3rd Interval
(I.2(I NAR lq "G
d-" 0.18
0.16 I) I E
1~0
..~ 0.12 RGR NAR
0.O~ RGR 18
O.O4 " " " " ,~-...--~ & 16 '-~
I I I II I I II I I
300 600 900 300 600 900 300 600 900
CO~ Concentration ( ~/ /-')
Fig. 2. RGR (AA), LAR (o,e) and NAR (o,.) for 'Bragg' soybeans grown in pots in 1981
(open symbols) and in the field in 1982 (closed symbols) in open-top field chambers at
elevated CO2 concentrations.
The first growth interval (Fig. 2) showed a small apparent decrease in LAR at
higher CO2 concentrations in both the 1982 seeded row crop study and the
1981 potted plant experiment. Bearing in mind that chamber placement
and CO2 dispensing in the 1982 field study did not begin until 8 days after
planting, it is perhaps not surprising that in the first interval there was
only a trend for increasing NAR and therefore (since RGR = LAR × NAR)
for increasing RGR with increasing CO2 concentration. In contrast, in the
1981 pot study, where treatments were imposed at seed planting, there
was a very marked effect of CO2 concentration on NAR and therefore on
RGR in the same time period.
125
In the 2nd interval, the RGR of the potted plants, which remained con-
stant across CO2 treatments, reflected about equally the influence of the
increasing NAR and decreasing LAR. In this interval, LAR was substantially
higher in the seeded row crop than in the pot study, although it responded
to CO2 treatments similarly to the potted plants. NAR, however, was not
only higher in the row crop than in the pot study, the response to CO2
continued at the highest treatment level, whereas in the pot study the
NAR no longer responded to CO2 concentration above 623 #l 1 -~. The RGR
of the row crop was, therefore, not only much higher than in the pot study
(0.17 vs. 0.11 at ambient CO2), but due to the contribution of NAR, it
increased with increasing CO2 concentration as well, reaching 0.20 at 945
#11 -l.
In the third interval, the influence of CO2 on NAR and RGR was no
longer observed in either study, although it was still apparent in the LAR
in both studies. The influence of elevated CO2 was qualitatively similar in
both growth conditions: lowered LAR and stimulated NAR and RGR
early in the season with decreasing effect as vegetative growth proceeded.
The major differences in growth observed in the field-grown soybeans,
for which root growth was presumably unrestricted, were (a) highly in-
creased LAR and (b) higher NAR values at high CO2 concentrations in the
2nd interval. In comparing growth of 'Bragg' soybeans under these two
systems, it is interesting that Sionit et al. (1984) found for soybeans during
pod filling an essentially linear leaf photosynthetic response to light up to
1600 ~E m -2 s -1 for field grown soybeans, whereas photosynthesis for
potted plants in neighboring chambers leveled off at 800 ~E m -~ s -1. The
relative effect of high CO2 was much greater for the potted plants.
Yield Response
The 1982--83 field experiments were conducted to directly address the
issue of elevated carbon dioxide effects on field crop growth, behavior
and yield. Field conditions were therefore maintained as closely as possible.
However, most work with CO2 effects on soybeans has been done in pots,
whether in open-top chambers or greenhouses in controlled environment
chambers, or in outdoor controlled environment chambers (SPAR units).
A survey of all CO2 soybean yield work for which growth conditions were
available was made. This work fell naturally into two classes: experiments
in which one very high CO2 concentration (> 1000/~l 1 -~) was compared
with a CO2 concentration near ambient (Table IV), and those in which
several CO2 levels were maintained, all between ambient and 1000/~l 1-1
(Table V). Only in the work of Sionit (Table V) was there a single elevated
CO2 treatment which was also less than 1000 #11 -~. The seed yield data
from these studies were regressed against CO2 concentration, and the yield
response ratios were calculated from predicted values at 1000 ~11-1/350
#11 -~ using a linear model (Table IV) and at 700~11-~/350 #11 -~ using either
a linear or quadratic model, as statistically appropriate (Table V).
126
Harvest index (HI) was either unaffected or decreased in all reports in
Tables IV and V, except those in which exposure was limited to later stages
of growth (Table IV, Experiments 4--6). These latter increases in HI due to
CO2 treatment during reproductive growth were also observed by Hardman
and Brun (1971). Soybean appears to be the only crop species for which
CO2 usually decreases HI (Cure, 1985). Increases in HI due to increased
CO2 concentration have been reported for barley (Gifford et al., 1973),
corn (Goudriaan and deRuiter, 1983; Rogers et al., 1983a) rice (Cock
and Yoshida, 1973; Yoshida, 1973) and wheat (Gifford, 1977, 1979; Sionit
et al., 1980, 1981; Goudriaan and deRuiter, 1983).
Further careful study of Tables IV and V shows no clear, substantial
effects of genotype or degree of determinacy on the response ratios. If
there is such an effect, it is overshadowed by the apparent effect of growth
conditions for both the 1000/350 gl 1-1 yield response ratio (Table IV,
cf entries 1 a--d, 2 a--b, 4--6 b) and the 700/350 ~1
1-1
yield response ratio
(Table V, cf entries 1 a--b, 2--4, 5--8 b and 9--10 b). The means from each
of the experiments were pooled to obtain an overall response at 700 ~l 1-1
of 1.29 + s.e.m. 0.11 and at 1000 ~11-1of 1.35 + s.e.m. 0.11.
Although we have no unequivocal evidence as to the effect of such field-
associated stresses as high leaf temperature on the photosynthetic or growth
response to elevated CO2, we now have evidence, accumulated mostly
from controlled environment experiments, that the growth response to
CO2 is dampened under conditions of nutrient stress for soybeans (Imai
and Murata, 1978; Sionit et al., 1981; Williams et al., 1981; Patterson and
Flint, 1982; Goudriaan and deRuiter, 1983; Sionit, 1983). Water stress
may be unique among commonly encountered field stresses because of the
positive effect of elevated CO2 on water-use efficiency for soybeans (Rogers
et al., 1983a; Valle et al., 1985) as well as other species (Carlson and Bazzaz,
1980). Thus, growth response of soybeans to CO2 should be enhanced in dry
conditions as has been shown for wheat (Gifford, 1979). The data from our
1982--83 field studies (this report) suggest that the effects of elevated CO2
on growth and yield of field-grown soybeans may be limited by the various
stresses associated with field conditions except in the presence of water
stress.
In order to predict with confidence how plants will respond to elevated
CO2 concentration in the field, we must first, make better use of controlled
environment facilities to explore interactions of environmental factors
(e.g. temperature, light, root restrictions on growth) and their impact on
the COx response, and second, characterize our field test facilities more
fully so as to understand the differences illustrated here both from year to
year and from site to site.
ACKNOWLEDGEMENT
Funding for this work was by the Department of Energy through inter-
agency No. DEAI-01-81-ERG0001 to the USDA.
127
REFERENCES
Ackerson, R.C., Havelka, U.D. and Boyle, M.G., 1984. CO:-enrichment effects of soybean
physiology If. Effects of stage-specific CO2 exposure. Crop Sci., 24: 1150--1154.
Acock, B., Baker, D.N., Reddy, V.R., McKinion, J.M., Whisler, F.D., Del Castillo, D. and
Hodges, H.F., 1982. Soybean responses to carbon dioxide: measurement and simulation
1981. 004 in series: Response of Vegetation to Carbon Dioxide. U.S. Dept. of Energy
and Agriculture.
Acock, B., Reddy, V.R., Del Castillo, D., Hodges, H.F., Baker, D.N., McKinion, J.M. and
Whisler, F.D., 1983. Soybean responses to carbon dioxide: measurement and simula-
tion
1982. 008 in series: Response of Vegetation to Carbon Dioxide. U.S. Dept. of
Energy and Agriculture.
Carlson, R.W. and Bazzaz, F.A., 1980. The effects of elevated CO2 concentrations on
growth, photosynthesis, transpiration and water use efficiency of plants. In: J.J.
Singh and A. Deepak (Editors), Environmental and Climatic Impact of Coal Utiliza-
tion. Academic Press, New York, pp. 608--623.
Cock, J.H. and Yoshida, S., 1973. Changing sink and source relations in rice (Oryza
sativa L.) using carbon dioxide enrichment in the field. Soil Sci. Plant Nutr., 19:
229--234.
Cooper, R.L. and Brun, W.A., 1967. Responses of soybeans to a carbon dioxide-en-
riched atmosphere. Crop Sci., 7: 455--457.
Cure, J.D., 1985. Carbon dioxide doubling response: a crop survey. In: B.R. Strain and
J.D. Cure (Editors), Direct Effects of Increasing Carbon Dioxide on Vegetation. U.S.
Dept. of Energy DOE/ER-0238, N.T.I.S., Springfield, VA.
Edwards, J.A., Reilly, J., Trabalka, J.R. and Reichle, D.E., 1984. An analysis of possible
future atmospheric retention of fossil fuel COs. DOE Report TR013, N.T.I.S., Dept.
of Commerce, Springfield, VA.
Gifford, R.M., 1977. Growth pattern, carbon dioxide exchange and dry weight distribu-
tion in wheat growing under differing photosynthetic environments. Aust. J. Plant
Physiol., 4: 99--110.
Gifford, R.M., 1979. Growth and yield of carbon dioxide enriched wheat under water-
limited conditions. Aust. J. Plant Physiol., 6: 367--378.
Gifford, R.M., Bremmer, P.M. and Jones, D.B., 1973. Assessing photosynthesis limitation
to grain yield in a field crop. Aust. J. Agric. Res., 24: 297--307.
Goudriaan, J. and deRuiter, H.E., 1983. Plant growth in response to COs enrichment,
at two levels of nitrogen and phosphorus supply. I. Dry matter, leaf area and develop-
ment. Neth. J. Agric. Sci., 31: 157--161.
Hardman, L.L. and Brun, W.A., 1971. Effect of atmospheric carbon dioxide enrichment
at different developmental stages on growth and yield components of soybeans.
Crop Sci., 11: 886--888.
Havelka, U.D. and Hardy, R.W.F., 1976. Legume N2 fixation as a problem in carbon
nutrition. In: W.E. Newton and C.J. Nyman (Editors), Proc. 1st Int. Symp. on Ni-
trogen Fixation. Washington State University Press, Pullman, Washington. pp. 456--
475.
Havelka, U.D., Ackerson, R.C., Boyle, M.G. and Wittenbach, V.A., 1984. COs-enrichment
effects on soybean physiology. I. Effects of long-term CO2 exposure. Crop Sci., 24:
1146--1150.
Imai, K. and Murata, Y., 1978. Effect of carbon dioxide concentration on growth and
dry matter production of crop plants. III. Relationship between COs concentration
and nitrogen nutrition in some C3- and C4-species. Japan J. Crop Sci., 47: 118--123.
Israel, D.W. and Rogers, H.H., 1982. The effect of N~-fixing ability of Rhizobium strain
on response of nodulated soybeans to atmospheric COs enrichment. In: Field Studies
of Plant Responses to Elevated Carbon Dioxide. 009 in series: Response of Vegetation
Carbon Dioxide. U.S. Dept. of Energy and Agriculture, pp. 122--161.
128
Jones, P., Allen, L.H., Jr., Jones, J.W., Boote, K.V. and Campbell, W.J., 1984. Soybean
canopy growth, photosynthesis and transpiration responses to whole-season carbon
dioxide enrichment. Agron. J., 76: 633--637.
Kimball, B.A., 1983. Ann. Report, U.S. Water Conservation Laboratory, Phoenix, AZ.
Kvet, J., Ondok, J.P., Necas, J. and Jarvis, P.G., 1971. Methods of growth analysis. In:
Z. Sestak, J. Catsky and P.G. Jarvis (Editors), Plant Photosynthetic Production Manual
of Methods. Dr. W. Junk, The Hague, pp. 343--391.
Lemon, E.R., (Editor) 1983. CO 2 and Plants. The response of plants to rising levels of
atmospheric carbon dioxide. A.A.A.S. selected symposia series. Westview Press,
Inc., Boulder, CO.
Maddox, J.J., 1974. Influence of supplemental CO2 on four southern varieties of soybean
(Glycine
max L. Merrill). Ph.D. Dissertation., Mississippi State University, MI, 121 pp.
Neter, J. and Wasserman, W., 1974. Applied Linear Statistical Models. Richard D. Irwin,
Homewood, IL.
Patterson, D.T. and Flint, E.P., 1982. Interacting effects of CO s and nutrient concentra-
tion. Weed Sci., 30: 389--394.
Rogers, H.H. (Coordinator), 1981. Field studies of plant responses to elevated carbon
dioxide levels. 005 in series: Response of Vegetation to Carbon Dioxide. U.S. Dept.
of Energy and Agriculture.
Rogers, H.H. and Bingham, G.E. (Coordinators), 1982. Field studies of plant responses
to elevated carbon dioxide. 009 in series: Response of Vegetation to Carbon Dioxide.
U.S. Dept. of Energy and Agriculture.
Rogers, H.H., Bingham, G.E., Cure, J.D., Smith, J.M. and Surano, K.A., 1983a. Responses
of selected plant species to elevated carbon dioxide in the field. J. Environ. Qual.,
12: 569--574.
Rogers, H.H., Heck, W.W. and Heagle, A.S., 1983b. A field technique for the study
of crop responses to elevated carbon dioxide concentrations. J. Air Pollut. Control
Assoc., 33: 42--44.
Rogers, H.H., Thomas, J.F. and Bingham, G.E., 1983c. Response of agronomic and forest
species to elevated atmospheric carbon dioxide. Science, 220: 428--429.
Rogers, H.H., Cure, J.D., Thomas, J.F. and Smith, J.M., 1984a. Influence of elevated
CO2 on growth of soybean plants. Crop Sci., 24: 361--366.
Rogers, H.H., Sionit, N., Cure, J.D., Smith, J.M. and Bingham, G.E., 1984b. Influence
of elevated carbon dioxide on water relations of soybeans. Plant Physiol., 74: 233--
238.
Sionit, N., 1983. Response of soybean to two levels of mineral nutrition in CO~-enriched
atmosphere. Crop Sci., 23: 329--333.
Sionit, N., Hellmers, H. and Strain, B.R., 1980. Growth and yield of wheat under CO 2
enrichment and water stress. Crop Sci., 20: 687--690.
Sionit, N., Mortensen, D.A., Strain, B.R. and Hellmers, H., 1981. Growth response of
wheat to CO 2 enrichment and different levels of mineral nutrition. Agron. J., 73:
1023--1027.
Sionit, N., Rogers, H.H., Bingham, G.E. and Strain, B.R., 1984. Photosynthesis and
stomatal conductance with CO2-enrichment of container- and field-grown soybean.
Agron. J., 76: 447--451.
Valle, R., Mishoe, J.W., Jones, J.W. and Allen, L.H., Jr., 1985. Transpiration rate and
water use efficiency of soybean leaves adapted to different CO 2 environments. Crop
Sci., 25: 477--482.
Williams, L.E., DeJong, T.M. and Phillips, D.A., 1981. Carbon and nitrogen limitations
on soybean seedling development. Plant Physiol., 68:1206--1209.
Yoshida, S., 1973. Effects of CO s enrichment at different stages of panicle development
on yield components and yield of rice
(Oryza sativa
L.). Soil Sci. Plant Nutr., 19:
311--316.
