Topical Review on Photosynthetic Energy Conversion Efficiency

Photosynthetic Energy Conversion Efficiency: Setting a

Baseline for Gauging Future Improvements in Important
Food and Biofuel Crops1
Rebecca A. Slattery and Donald R. Ort*
Department of Plant Biology (R.A.S., D.R.O.), Institute for Genomic Biology (R.A.S., D.R.O.), and Global
Change and Photosynthesis Research Unit (D.R.O.), United States Department of Agriculture, Urbana,
Illinois 61801

Downloaded from https://academic.oup.com/plphys/article/168/2/383/6113608 by guest on 06 January 2024

ORCID IDs: 0000-0002-7165-906X (R.A.S.); 0000-0002-5435-4387 (D.R.O.).

The conversion efficiency («c) of absorbed radiation into biomass (MJ of dry matter per MJ of absorbed photosynthetically active
radiation) is a component of yield potential that has been estimated at less than half the theoretical maximum. Various strategies
have been proposed to improve «c, but a statistical analysis to establish baseline «c levels across different crop functional types is
lacking. Data from 164 published «c studies conducted in relatively unstressed growth conditions were used to determine the
means, greatest contributors to variation, and genetic trends in «c across important food and biofuel crop species. «c was greatest
in biofuel crops (0.049–0.066), followed by C4 food crops (0.046–0.049), C3 nonlegumes (0.036–0.041), and finally C3 legumes
(0.028–0.035). Despite confining our analysis to relatively unstressed growth conditions, total incident solar radiation and
average growing season temperature most often accounted for the largest portion of «c variability. Genetic improvements in
«c, when present, were less than 0.7% per year, revealing the unrealized potential of improving «c as a promising contributing
strategy to meet projected future agricultural demand.

Substantial increases in yield are needed to feed and mays), with similar concerns that yield trends may also
fuel the world’s growing human population. With an be decreasing in some major growing regions (Lobell
estimated population of nine billion people by the and Gourdji, 2012; Ray et al., 2013).
middle of this century (Lutz and Samir, 2010) and rising Efforts to increase yields in the next few decades must
affluence resulting in greater consumption of grain-fed also account for environmental and sustainability goals
animal products (Cirera and Masset, 2010), different (Sayer et al., 2013) as well as heightened environmental
studies predict that, by midcentury, global crop pro- stresses predicted to occur due to climate change, which
duction will need to increase 60% to 120% over 2005 are already responsible for some of the stagnation in
levels without the expansion of agricultural land area yield increases. Anthropogenic sources of greenhouse
(Tilman et al., 2011; Alexandratos and Bruinsma, 2012). gases have caused an approximately 1°C increase in land
Doubling yields in major food and fuel crops requires surface temperatures since 1900, and global mean sur-
considerable effort, especially as yields are beginning to face temperatures are likely to increase by up to 2.4°C to
plateau in many major food crops. Yield increases nec- 4.8°C by the end of the century (IPCC, 2013). Drought is
essary for doubling productivity by midcentury are es- also expected to become more frequent and intense in
timated at 1.16% to 1.31% each year in all cereals (Hall many regions of the world (Dai, 2011; IPCC, 2013). Of
and Richards, 2013), 1.7% per year in wheat (Triticum the variability present in major food crop yield gains,
aestivum; Rosegrant and Agcaoili, 2010), and 2.4% 30% can be explained by climate change alone (Lobell
(noncompounding average per year) across all major and Field, 2007), with drastic decreases in barley (Hor-
grain crops (Ray et al., 2013). However, global mean deum vulgare), maize, rice, sorghum (Sorghum bicolor),
increases from the past 20 to 30 years suggest that yield soybean, and wheat yields as average growing season
gains in rice (Oryza sativa) and wheat are approximately temperatures surpass the temperature optimum for each
1% (Lopes et al., 2012; Manès et al, 2012; Ray et al., 2013) crop (Lobell and Gourdji, 2012). Current levels of at-
and declining in some areas of the world (Cassman et al., mospheric CO2 concentration [CO2] are the highest they
2010; Fischer and Edmeades, 2010; Long and Ort, 2010; have been in at least 800,000 years (IPCC, 2013). Elevated
Ray et al., 2013). Global yearly increases are estimated at [CO2] increases water use efficiency (Ainsworth and
1.3% in soybean (Glycine max) and 1.6% in maize (Zea Long, 2005, Bernacchi et al., 2007, Leakey et al., 2009), but
probably not to an extent that would mitigate the
resulting reductions in yield caused by higher tempera-
1
This work was supported by the U.S. Department of Agriculture
ture and higher vapor pressure deficit (Ort and Long,
Agricultural Research Service. 2014). Additionally, any fertilization effects on C3 yields
* Address correspondence to d-ort@illinois.edu. due to elevated [CO2] would be at least in part negated by
www.plantphysiol.org/cgi/doi/10.1104/pp.15.00066 drought and temperature stress, leaving yield increases

Plant PhysiologyÒ, June 2015, Vol. 168, pp. 383–392, www.plantphysiol.org Ó 2015 American Society of Plant Biologists. All Rights Reserved. 383
Slattery and Ort

far from optimal (Long et al., 2006a; Lobell and Gourdji, decades, the extent to which genetic improvements ver-
2012). sus climate change have contributed to changes in «c can
also be assessed in individual crop species.
These analyses used primary literature to compare the
mean «c among and within several food and biofuel crop
USING PAST LITERATURE TO BETTER
species (Table I). Additionally, the relationships between
UNDERSTAND THE ROLE THAT IMPROVEMENTS
«c and environmental and genetic variables were ex-
IN CONVERSION EFFICIENCY HAVE PLAYED IN
amined over several decades within major food crops.
INCREASED YIELDS
Briefly, studies containing «c (also referred to as radia-
To double yields in less than 50 years with the addi- tion use efficiency) measurements under relatively un-
tional challenges of climate change, research needs to stressed growing conditions were collected. «c values were
target yield-related processes that have potential for extracted from the resulting 164 studies (Supplemental
Table S1). «c is generally calculated as the slope of crop

Downloaded from https://academic.oup.com/plphys/article/168/2/383/6113608 by guest on 06 January 2024

considerable improvement. The theoretical maximum
conversion efficiency («c) of photosynthetically active accumulated biomass (in terms of mass or energy) versus
radiation (PAR) into plant biomass has been calculated intercepted or absorbed solar energy or PAR by the
in C3 (0.094) and C4 (0.123) plants in optimal conditions. canopy. For the purpose of these analyses, all values
The calculations are based on the minimal amounts of were standardized to units of MJ of dry matter per MJ of
energy that could be lost due to various steps in trans- absorbed PAR before statistically testing differences at
forming intercepted PAR into plant biomass. The steps a = 0.1. An additional aim was to estimate rates of gain in
where energy is lost include light reflection and trans- «c due to breeding and to determine whether «c variation
mission, photochemical inefficiency resulting from ex- in major food crops over time was more significantly
cess energy in absorbed blue light, thermodynamic associated with breeding or climate variables. When
limitations, carbohydrate synthesis, photorespiration (C3 available, crop information and growing conditions from
only), and respiration (Zhu et al., 2008, 2010). Estimated each study were used as independent variables in mul-
at less than half of the theoretical maximum in C3 and C4 tiple regression analyses. These variables included year of
plants in optimal conditions, «c appears to be an ideal release (YOR), mean annual [CO2] for the years that the
candidate for increasing yield potential (Beadle and experiments were conducted, mean growing season
Long, 1985; Zhu et al., 2010). Moreover, considerable temperature (T), available incident solar radiation during
variation is present in «c, as it is sensitive to greenhouse the growing season (St), the amount of precipitation (rain
gases and weather-related variables predicted to inten- and irrigation) available during the growing season, and
sify due to climate change (Sinclair and Muchow, 1999; plant density (maize only). Varieties included in the
Slattery et al., 2013). Because «c is seemingly well below analyses were indicated for each crop and subgroup
its theoretical maximum and is a highly variable pa- (Supplemental Table S2). In food crops with significant
rameter across growing environments and crop species, positive correlations between «c and YOR, the regression
potential methods to improve «c have been identified coefficient was used to determine the time to double or
and reviewed. The most general strategy for improving reach the maximum «c, assuming no major changes in
«c at the leaf level involves improving the efficiency of trends due to genetic or environmental factors. (For a
carboxylation by Rubisco while limiting oxygenation in fully detailed explanation of the methods used in this
C3 plants (Zhu et al., 2010; Parry et al., 2011; Raines, study, see Supplemental Materials and Methods S1.)
2011; Ainsworth et al., 2012; Evans, 2013). At the canopy
level, targets include using altered canopy architecture
and antenna size to improve light distribution in dense-
STATUS OF «C IN MAJOR FOOD CROP SPECIES
canopy crops (Zhu et al., 2010; Ort et al., 2011; Parry
et al., 2011; Ainsworth et al., 2012; Reynolds et al., 2012), Previous estimates putting «c means in food crops at
maximizing nitrogen partitioning, and enhancing spike approximately one-third to one-half of the maximum
photosynthesis (Reynolds et al., 2012). (Beadle and Long, 1985; Zhu et al., 2010) were consistent
While there is seemingly substantial potential for in- with the results from this study, with the exception of
creasing «c in major food and biofuel crops, judging the many legumes demonstrating values below one-third
effectiveness of each strategy is difficult without baseline the C3 maximum. Maize, a highly developed and in-
estimates of «c and rates of gain to date in individual tensively grown crop, had the greatest mean «c among
crops. Sensitivity to environmental conditions implies food crops included in the analysis (0.0488; Fig. 1; Table
that using single studies may not be the best method for II) but was still less than one-half the predicted maxi-
gauging the status of «c within individual crop species. mum of 0.123 (Zhu et al., 2010). Grain sorghum was
Therefore, a meta study of the large body of literature slightly lower than maize (0.0455; P = 0.10; Fig. 1) and
that exists on «c should provide insight into the current was only 37% of the maximum. C3 nonlegume crop
status of «c in individual crops, the extent that «c varies means were significantly lower than C4 means and
among food and biofuel crop species, which crops ranged from 40% to 45% of the predicted maximum,
demonstrate greater potential for «c improvements, and with the greatest C3 mean in potato (Solanum tuberosum;
inherent characteristics that may be benefitting crops 0.0414; Fig. 1; Table II). Except for peanut (Arachis
with greater realized «c. Since the literature spans several hypogaea), which had a mean «c of 0.0346 (Fig. 1; Table II)
384 Plant Physiol. Vol. 168, 2015
 Means and Trends in Crop Energy Conversion Efficiency

Table I. Important food and C4 biofuel crop species used in «c analyses

Species further divided into genetic components are indicated. Biomass energy content for vegetative (V) and combined vegetative and repro-
ductive (V+R) stages used for converting «c values to energy units are indicated for each crop.
Energy
Common Food or Energy Groups by Cultivar, Content
Species Type Energy Content Data Source
Name Crop Species, or Hybrid
V V+R
MJ kg21
Z. mays Maize C4 Botha 17.0 17.0 Penning de Vries et al. (1989)
S. bicolor Sorghum C4 Botha Energy/biomass/forage grain 17.6 17.3 Amthor et al. (1994)
S. tuberosum Potato C3 Food 17.0 15.8 Penning de Vries et al. (1989)
O. sativa Rice C3 Food New hybrids 15.1 15.9 Penning de Vries et al. (1989)
indica

Downloaded from https://academic.oup.com/plphys/article/168/2/383/6113608 by guest on 06 January 2024

japonica
Basmati
T. aestivum Wheat C3 Food Spring 17.0 16.6 Penning de Vries et al. (1989)
Winter
H. vulgare Barley C3 Food 16.1 15.6 McKendry (2002; V); Sinha et al.
(1982; V+R)
A. hypogaea Peanut C3 Food 17.9 23.3 Penning de Vries et al. (1989)
G. max Soybean C3 Food 18.1 19.8 Amthor et al. (1994)
Cicer arietinum Chickpea C3 Food 17.9 18.6 Penning de Vries et al. (1989)
Cajanus cajan Pigeonpea C3 Food 17.9 18.4 Penning de Vries et al. (1989)
P. virgatum Switchgrass C4 Energy 17.4 17.4 McKendry (2002)
Saccharum spp. Sugarcane C4 Energy 17.4 17.4 Botha (2009)
Miscanthus spp. Miscanthus C4 Energy M. giganteus 18.5 18.5 McKendry (2002)
M. sinensis
a
Maize studies completely overlapped from food to energy analyses. Sorghum food and energy cultivars were separated and analyzed in the
respective analyses.

and was approximately 38% of the maximum, all le- Additionally, legume «c may have been affected by
gume means were approximately 0.028 and 31% of the nitrogen fixation, the costs of which have been deter-
possible maximum for C3 crops (Fig. 1; Table II). mined and vary by study. One study reports a 5% greater
One caveat is that this study omitted values from the photon energy requirement for nitrogen fixation com-
literature that included belowground biomass, with the pared with the combined cost of NH4+ and NO32 as-
exception of peanut and potato, because (1) studies similation that occurs in most nonlegume crops (Andrews
basing «c measurements on total aboveground and be- et al., 2009). In terms of carbon usage, the proportion of
lowground biomass were minimal and (2) belowground
harvesting methods differed greatly and, therefore, may
have skewed the results. «c is estimated to increase by
10% to 20% when accounting for belowground biomass
in annual plants (Sinclair and Muchow, 1999), which
would result in an approximate increase of 0.01 in «c in
food crops from this analysis, but this still would not
account for the large disparity between measured and
theoretical values. The greatest C3 «c was in potato,
which included belowground biomass but was still only
44% of the theoretical maximum. However, the omission
of belowground biomass in the calculation of «c could
have contributed to the disproportionately lower «c in
legume crops if belowground biomass energy is greater
in legumes compared with other crops. Indeed, below-
ground biomass (roots and nodules) of soybean contains
more energy (18.3 MJ kg21) than sorghum belowground
tissue (16.7 MJ kg21; Amthor et al., 1994). Based on en-
ergy contents reported on a per area basis by Amthor
et al. (1994), accounting for soybean belowground bio-
mass would increase whole-plant energy content by al-
most 6%. However, this would only increase soybean «c Figure 1. Calculated «c means in 10 major food crops. Crops are or-
to approximately 0.03, which is still well below the «c of ganized by C4, C3 nonlegume, and C3 legume categories. Sample size is
C3 nonlegume crops (Fig. 1; Table II). shown on the right axis. Error bars represent 90% confidence intervals.

Plant Physiol. Vol. 168, 2015 385

Slattery and Ort

Table II. Numerical data from mean «c analyses in major food crops
Species, common name, photosynthetic type, and any groups or specifications within species are in-
dicated. «c means and SE are reported along with sample size (n) for each species and group within species.
Groups by Cultivar, Species,
Species Common Name Type Mean «c SE n
or Hybrid
Z. mays Maize C4 0.0488 0.001 194
S. bicolor Sorghum C4 Grain type only 0.0455 0.002 44
S. tuberosum Potato C3 0.0414 0.002 48
O. sativa Rice C3 0.0399 0.001 132
New hybrids 0.0472 0.002 29
indica 0.0442 0.002 25
japonica 0.0388 0.002 57

Downloaded from https://academic.oup.com/plphys/article/168/2/383/6113608 by guest on 06 January 2024

Basmati 0.0273 0.003 21
T. aestivum Wheat C3 0.0378 0.001 242
Spring 0.0352 0.001 105
Winter 0.0399 0.001 137
H. vulgare Barley C3 0.0364 0.002 49
A. hypogaea Peanut C3 0.0346 0.002 60
G. max Soybean C3 0.0282 0.001 159
C. arietinum Chickpea C3 0.0283 0.002 52
C. cajan Pigeonpea C3 0.0280 0.002 40

assimilated carbon that is diverted to the nodules for (Supplemental Table S3). At approximately one-half
nitrogen fixation is reported as 7% to 19% (Gordon of the C3 theoretical maximum, the «c of new hybrid
et al., 1987; Vessey et al., 1988; Hansen et al., 1992, (0.0472) and indica (0.0442) rice varieties was significantly
1993; Fujikake et al., 2003; Ito et al., 2006). Correcting greater than that of japonica (0.0388) and basmati rice
for these costs on «c would result in a range of 0.03 to (0.0273; Fig. 2). New hybrid rice «c was not significantly
0.034 for the legumes in this study (excluding pea- different from maize «c (P = 0.51), and neither new hy-
nut). These values are still lower than nonlegume «c brid nor indica «c was significantly different from grain
values (Fig. 1; Table II), but other factors related to sorghum «c (P = 0.49 and P = 0.72, respectively). How-
nitrogen fixation may also limit legumes, such as the ever, a negative relationship between available St and
delay in forming mature nodules early in the growing subgroup «c suggested that the significantly greater «c in
season (Andrews et al., 2009). Therefore, increasing new hybrid and indica varieties may be associated with
the efficiency of nitrogen fixation may represent an growth conditions rather than genetic enhancements
additional means to improve «c in legumes that has (Supplemental Table S3). Significant differences were
previously received little attention with regard to in- also evident between the «c means of spring wheat
creasing photosynthetic efficiency. (0.0352) and winter wheat (0.0399; P , 0.01; Fig. 2; Table
Peanut was the anomaly within the legume group, II). However, mean St and T were once again lower in
with «c more similar to nonlegume C3 crops than to
legumes. Although a portion of belowground bio-
mass was included in the analyses of all peanut
studies included, this only comprised the fruiting
bodies growing underground and did not include
the rest of the root biomass. A more likely explana-
tion for the disparity between peanut and other le-
gumes in this study was the difference in reported
energy contents. Peanut whole-plant energy content
was 1.2 times greater than that in the rest of the le-
gumes (Table I), which corresponded to an approx-
imately 1.2 times greater efficiency (Fig. 1; Table II).

«C VARIES AMONG GROUPS WITHIN RICE AND

WHEAT, BUT THIS MAY BE THE RESULT OF
VARYING GROWTH CONDITIONS
Significant differences in «c were evident within the Figure 2. Calculated «c means for categories within rice (top) and
subgroups of rice and wheat (Fig. 2) but may have been wheat (bottom). Sample size is shown on the right axis. Error bars
confounded by differences in growth environments represent 90% confidence intervals.

386 Plant Physiol. Vol. 168, 2015

 Means and Trends in Crop Energy Conversion Efficiency

winter wheat compared with spring wheat (Supplemental giant reed (Arundo donax) and the C4 annual energy crop
Table S3), the effects of which are discussed below. sorghum supported this point. «c in giant reed, which
demonstrates photosynthetic rates typical of C3 plants
(Balota et al., 2008), was greater than the «c of the C4
«Cs IN BIOENERGY CROPS annual sweet sorghum (Ceotto et al., 2013). However,
the intercept of the relationship between aboveground
Not surprisingly, C4 crop «c was almost always biomass and intercepted radiation in giant reed was
greater than C3 «c due to inherent properties of C4 greater than zero, suggesting that rhizome energy con-
photosynthesis. C4 plants concentrate CO2 at the site of tributions to aboveground biomass were increasing
carboxylation, thereby inhibiting energy losses to pho- shoot growth rates independent of radiation absorption
torespiration and increasing the maximum potential «c (Ceotto et al., 2013).
as compared with C3 plants (Zhu et al., 2008, 2010).
However, «c was often greater in C4 energy crops

Downloaded from https://academic.oup.com/plphys/article/168/2/383/6113608 by guest on 06 January 2024

compared with C4 food crops. At 0.066, nongrain sor-
ghum varieties had the greatest «c mean of the energy TRENDS IN MAJOR FOOD CROP «C OVER THE PAST
crops included and reached 54% of the theoretical mean FEW DECADES
for C4 crops (Fig. 3; Table III). Switchgrass (Panicum Prior to conducting multiple regression analyses, the
virgatum) «c was not significantly lower than sorghum «c individual relationships between «c and independent
(P = 0.44) and was also greater than 50% of theoretical variables were determined in six major food crops (Fig.
(0.0618; Fig. 3; Table III). Sugarcane (Saccharum officina- 4). In multiple regression analyses, the best model was
rum) was similar to sorghum (P = 0.16) and switchgrass selected using the lowest corrected Akaike information
(P = 0.42), with a mean of 0.0581 (Fig. 3; Table III), and criterion (AICc) score, and inclusion of an independent
was not significantly different from the Miscanthus 3 variable in the model demonstrated that the variable
giganteus mean of 0.0503 (P = 0.11; Fig. 3; Table III). accounted for a large portion of the variability in «c for
Maize «c was the second lowest of the bioenergy crops that crop. The variables included in the final model
(0.0488; Fig. 3; Table III), despite having the highest «c of closely matched corresponding partial correlation co-
the food crops (Fig. 1; Table II). Miscanthus sinensis had efficients, demonstrating that colinearity between in-
the lowest mean of the bioenergy crops (0.027) and was dependent variables was not a significant factor in the
only 23% of the maximum theoretical for C4 crops (Fig. multiple regression results (Supplemental Table S4).
3; Table III). The most commonly included variables were T and St,
The apparent disparity between C4 food and bio- which were included in five linear multiple regression
energy crop «c may be the result of plant growth habit. models each (Table IV). YOR was only included in four
Unlike the annual C4 food crops, most of the C4 energy models, and [CO2] only in one model (Table IV). The
crops were perennial grasses, which are expected to AICc order of single variables included in the final
demonstrate greater aboveground biomass than annuals model indicated relative correlation strengths for
early in the season. Once established, perennials such as each variable. T was the first variable included in
M. giganteus and switchgrass draw upon belowground three of the models (peanut, soybean, and wheat), St
reserves from the previous season to facilitate growth in two of the models (rice and sorghum), and YOR
after emergence that is independent of absorbed radia- in one model (maize; Table IV). Although density
tion (Dohleman et al., 2012). This would inflate «c in was also included in the maize analyses, there was
perennial C4 biofuel crops during early crop growth. A no significant correlation between maize «c and
comparison by Ceotto et al. (2013) of the C3 perennial density, even though selection for density tolerance
is shown to correlate with increasing yield (Duvick,
2005). Additionally, substituting density for YOR
did not result in a significant correlation between
density and «c. Due to reduced sample size, pre-
cipitation available during the growing season was
only included in the analyses for soybean but was
not in the final selected model.
When St was included in a model, the simple regres-
sion coefficient was always negative (Table IV). Negative
correlations of «c with St within several food crops rein-
forced the notion that available light in excess of photo-
synthetic capacity decreases «c (Sinclair and Muchow,
1999; Slattery et al., 2013). Lowering St to a point near
light saturation at the top of the canopy increases «c but
ultimately depresses yield potential, due to less overall
energy available to the crop. Therefore, altering pigment
Figure 3. Calculated «c means in major C4 biofuel crops. Sample size is concentrations or canopy architecture in canopies with
shown on the right axis. Error bars represent 90% confidence intervals. high leaf area indices can optimize light availability
Plant Physiol. Vol. 168, 2015 387
Slattery and Ort

Table III. Numerical data from mean «c analyses in C4 bioenergy crops

Species, common name, photosynthetic type, and any groups or specifications within species are in-
dicated. «c means and SE are reported along with sample size (n) for each species and group within species.
Species Common Name Type Groups by Cultivar, Species, or Hybrid Mean «c SE n
Z. mays Maize C4 0.0488 0.001 194
S. bicolor Sorghum C4 Energy, biomass, forage 0.0660 0.005 12
P. virgatum Switchgrass C4 0.0618 0.003 26
Saccharum spp. Sugarcane C4 0.0581 0.003 22
Miscanthus spp. Miscanthus C4 M. giganteus 0.0503 0.004 20
M. sinensis 0.0279 0.004 16

Improving «c stress tolerance is becoming increas-

Downloaded from https://academic.oup.com/plphys/article/168/2/383/6113608 by guest on 06 January 2024

among leaf layers and decrease wasted incident radia-
tion (Long et al., 2006b; Ort et al., 2011; Drewry et al., ingly important, as current work to increase yields is
2014). This should have been evident in new rice varie- making crops more sensitive to detrimental climate
ties bred for reduced tillering, as fewer tillers was hy- change effects. Breeding for greater yields in optimal
pothesized to alter canopy structure, allowing greater conditions has resulted in greater sensitivity to the
penetration of light in the canopy and, therefore, environment (i.e. greater yield instability in less fa-
greater «c (Peng et al., 2008). In fact, «c was greatest in vorable conditions) in maize (Lobell et al., 2014) and
the new hybrid varieties (Table II; Fig. 2). However, soybean (Koester et al., 2014; Rincker et al., 2014).
these new hybrid varieties were also grown in rela- This may explain the lack of significant correlations
tively dim conditions as compared with the other between «c and density in maize. Although «c was
subgroups (Supplemental Table S3), preventing any expected to increase with density, a greater sensitivity
conclusions regarding the effectiveness to this ap- to temperature and, therefore, vapor pressure deficit
proach in conditions where incident light oversatu- could negate those benefits (Lobell et al., 2014). It is
rates photosynthetic capacity. Nonetheless, frequent also more difficult for newer, high-yielding cultivars
negative relationships between «c and St in food crops of wheat to realize maximum yields in the field as
present a solid argument for improving light distri- T stress becomes more common (Gourdji et al., 2013).
bution and use in dense food crop canopies. Since «c is sensitive to the environment (Table IV;
Multiple regression analyses indicated that, in ad- Sinclair and Muchow, 1999; Slattery et al., 2013), im-
dition to St, T accounted for the greatest proportion proving stress tolerance deserves even greater atten-
of the variation in «c in most food crops that were tion going forward in order to mitigate the negative
otherwise classified as experiencing optimal growth effects of climate change on yield.
conditions. T was negatively correlated with rice,
wheat, and soybean «c (Table IV). Positive correla-
tions were evident between «c and T in maize and
PROJECTED TIME TO DOUBLE CURRENT «C VALUES
peanut (Table IV), but as temperatures continue to
AND REACH THEORETICAL MAXIMUM «C IN
rise with predicted changes in climate, most likely all
MAJOR FOOD CROPS
food crops will begin to suffer decreases in «c. Even in
peanut, a crop where increases in T were positively Positive correlation of «c with YOR was limited across
correlated with «c in this analysis, recent studies have food crops and only demonstrated rates of increase of
found that expected increases in T will result in de- less than 0.7% per year. Relationships between «c and
creases in photosynthesis that will not be alleviated YOR determined in the absence of environmental var-
by elevated [CO2] (Prasad et al., 2003). A similar re- iability in wheat (Shearman et al., 2005; Sadras et al.,
sult was found in soybean grown in the field under 2012) and soybean (Koester et al., 2014) were reported
elevated [CO2] and elevated T, where elevated [CO2] as less than 0.65% gain per year in «c and were con-
had little effect on photosynthesis when T reached sistent with the rates from this study. At best, these
above optimal (Ruiz-Vera et al., 2013). While in- rates are half of the rates of yield increase necessary to
creasing temperatures can negatively affect crop double crop production by midcentury (Rosegrant and
growth and development in many ways, specific in- Agcaoili, 2010; Hall and Richards, 2013; Ray et al.,
hibitions to photosynthesis at the leaf level include de- 2013). Consequently, projections of when «c would
creased Rubisco specificity to CO2, limited ribulose double or reach the theoretical maximum suggest that
1,5-bisphosphate regeneration, and destabilization of the rate of genetic advancements to the present are not
Rubisco activase, which can have severe implications enough to double «c by the middle of the century in the
on C3 and even C4 photosynthesis at very high but crops studied. In maize, the food crop with the greatest
increasingly frequent temperatures (for review, see mean (Fig. 1) and greatest rate of increase with YOR
Ainsworth and Ort, 2010). Therefore, mitigating these (Table V), «c would not double until the year 2134, and
harmful effects on leaf photosynthesis through trans- the maximum would be reached approximately 70
genic approaches should be a priority along with im- years later (Table V). In peanut, the estimated time for
proving leaf photosynthetic efficiency. «c to double was approximately 250 years, while
388 Plant Physiol. Vol. 168, 2015
 Means and Trends in Crop Energy Conversion Efficiency

Downloaded from https://academic.oup.com/plphys/article/168/2/383/6113608 by guest on 06 January 2024

Figure 4. Relationships between «c and individual independent variables in six major food crops. Independent variables in-
cluded YOR, mean annual [CO2] during the measurement period, T, available St, and water available during the growing season
as precipitation and irrigation (H2O). Lines represent least-squares regression, with corresponding significance levels and
correlation coefficients in each graph. «c versus density in maize is not shown.

reaching the maximum would not occur for at least 400 would take approximately one millennium (Table V).
years (Table V). The year of doubling «c in wheat was These projections, based on trends in «c spanning sev-
2357, whereas reaching the maximum would occur in eral decades, demonstrated that breeding and biotech-
2391 (Table V). Due to having the low absolute value of nology to date have not necessarily selected for
«c and the lowest rate of gain in soybean, doubling «c increasing «c as a high priority. Thus, there is potentially
Plant Physiol. Vol. 168, 2015 389
Slattery and Ort

Table IV. Multiple regression analyses of «c in six major food crops

The final model and ranking of variables included in the final model were determined using the lowest
AICc. Variable coefficients and significance were determined using the final model.
Crop Variable Rank Modela AICc Coefficient (1023) P
Peanut (n = 51) YOR, St, T 2540.1 ,0.001
1 T 2536.0 0.426 ,0.001
2 YOR 2531.8 0.123 ,0.05
3 St 2526.6 20.00238 0.12
Soybean (n = 117) YOR, [CO2], St, T 21,270.2 ,0.0001
1 T 21,231.8 21.04 ,0.0001
2 St 21,226.3 20.00824 ,0.0001
3 [CO2] 21,204.3 0.0524 ,0.05

Downloaded from https://academic.oup.com/plphys/article/168/2/383/6113608 by guest on 06 January 2024

4 YOR 21,201.3 0.0293 0.21
Rice (n = 102) St, T 21,133.8 ,0.0001
1 St 21,125.0 20.00960 ,0.0001
2 T 21,014.3 20.729 ,0.01
Wheat (n = 159) YOR, St, T 21,451.2 ,0.0001
1 T 21,447.9 20.746 ,0.001
2 YOR 21,438.4 0.105 ,0.05
3 St 21,437.8 20.00284 0.13
Sorghum (n = 23) 1 St 2228.3 20.00739 0.11
Maize (n = 149) YOR, T 21,347.8 ,0.0001
1 YOR 21,346.0 0.346 ,0.0001
2 T 21,344.0 0.521 ,0.01
a
Independent variables included YOR, mean annual [CO2] during the measurement period, T, and
available St. Water available as precipitation and irrigation was included when sample size changed by
less than 10% after including it in the analyses. Density was included in maize analyses but was not in the
selected model.

a large amount of room for improvement in this key has room for improvement and is receiving in-
factor and, therefore, yield potential. creasing amounts of attention is promising. Targets
for improving «c in various manners have already
been identified and reviewed (Amthor, 2010, Zhu
CONCLUSION et al., 2010; Parry et al., 2011; Raines, 2011; Ainsworth
et al., 2012; Reynolds et al., 2012; Evans, 2013) and
As greater increases in yields are needed to feed and have the potential to greatly alter the current trends
fuel the world’s population, targets such as «c are key in «c improvement. This study emphasizes the im-
to reaching these goals. This assessment aimed to de- portance of using strategies that improve nitrogen
termine the mean «c in several important food and fixation efficiency in legumes, canopy light distri-
biofuel crops, test the key contributors to variation in bution, and tolerance to higher temperatures to
«c, and determine genetic trends in «c. As expected, increase genetic gains and limit detrimental envi-
mean «c values in food crops were greatest in C4, fol- ronmental effects on «c . As these strategies are
lowed by nonlegume C3, and were lowest in legume implemented to improve «c and, therefore, yield
C3 plants. All food crop means were lower than one- potential, these «c means and trends will serve as
half the theoretical maximum. Bioenergy crop «c means a baseline to track the relative success of each
were much greater than those in food crops, and some, approach.
including the energy crops sorghum and switchgrass,
exceeded 50% of the maximum for C4 grasses. However,
«c values for perennial grasses may have been inflated if
measured during the growing season interval when Table V. Summary of «c trends and projections in major food crops
stored rhizome reserves are mobilized and contribute to Positive trends in «c from multiple regression analyses were used to
aboveground biomass accumulation. Reported variation project the year in which «c will double and reach the theoretical
in «c was found to be generally negatively correlated maximum, assuming no changes in the coefficients due to climate
change, breeding intensity, etc.
with St and T. Positive correlations with YOR were only
present in a few food crops, and rates of increase were Crop YOR Slope Year of Doubling Year of Maximum
relatively low, suggesting that «c will not double in most year 21
3 10 23

crops before the middle of the century at the current rate Maize 0.346 2134 2176
of increase. Wheat 0.105 2357 2391
While these findings show that there has been Peanut 0.123 2273 2317
little progress to date in improving «c, the fact that «c Soybean 0.029 2966 2986

390 Plant Physiol. Vol. 168, 2015

 Means and Trends in Crop Energy Conversion Efficiency

Supplemental Data Drewry DT, Kumar P, Long SP (2014) Simultaneous improvement in

productivity, water use, and albedo through crop structural modifica-
The following supplemental materials are available. tion. Glob Change Biol 20: 1955–1967
Duvick DN (2005) Genetic progress in yield of United States maize (Zea
Supplemental Table S1. Data sources used in analyses.
mays L.). Maydica 50: 193–202
Supplemental Table S2. Varieties included in crop and subgroup analyses. Evans JR (2013) Improving photosynthesis. Plant Physiol 162: 1780–1793
Fischer RA, Edmeades GO (2010) Breeding and cereal yield progress. Crop
Supplemental Table S3. Means and ranges of independent variables used Sci 50: S-85–S-98
in analyses. Fujikake H, Yamazaki A, Ohtake N, Sueyoshi K, Matsuhashi S, Ito T,
Supplemental Table S4. Partial correlation coefficients from regression Mizuniwa C, Kume T, Hashimoto S, Ishioka NS, et al (2003) Quick and
analyses. reversible inhibition of soybean root nodule growth by nitrate involves a
decrease in sucrose supply to nodules. J Exp Bot 54: 1379–1388
Supplemental Materials and Methods S1. Detailed description of data Gordon AJ, Mitchell DF, Ryle GJA, Powell CE (1987) Diurnal production
collection, manipulation, and analysis. and utilization of photosynthate in nodulated white clover. J Exp Bot 38:
84–98

Downloaded from https://academic.oup.com/plphys/article/168/2/383/6113608 by guest on 06 January 2024

Gourdji SM, Mathews KL, Reynolds M, Crossa J, Lobell DB (2013) An
assessment of wheat yield sensitivity and breeding gains in hot envi-
ACKNOWLEDGMENTS ronments. Proc R Soc B Biol Sci 280: 1–8
We thank Dr. Adam Davis for providing advice on the statistical analyses Hall AJ, Richards RA (2013) Prognosis for genetic improvement of yield
used in this study and Dr. Elizabeth Ainsworth for advice on the meta- potential and water-limited yield of major grain crops. Field Crops Res
analysis procedures. 143: 18–33
Hansen AP, Yoneyama T, Kouchi H (1992) Short-term nitrate effects on
Received January 16, 2015; accepted March 25, 2015; published March 31, hydroponically-grown soybean cv. Bragg and its supernodulating mu-
2015. tant. J Exp Bot 43: 9–14
Hansen AP, Yoneyama T, Kouchi H, Martin P (1993) Respiration and ni-
trogen fixation of hydroponically cultured Phaseolus vulgaris L. cv. OAC
LITERATURE CITED Rico and a supernodulating mutant. Planta 1993: 538–545
IPCC (2013) Summary for policymakers. In TF Stocker, D Qin, GK Plattner,
Ainsworth EA, Long SP (2005) What have we learned from 15 years of free-
M Tignor, SK Allen, J Boschung, A Nauels, Y Xia, V Bex, PM Midgley,
air CO2 enrichment (FACE)? A meta-analytic review of the responses of
eds, Climate Change 2013: The Physical Science Basis. Contribution of
photosynthesis, canopy properties and plant production to rising CO2.
Working Group I to the Fifth Assessment Report of the Intergovern-
New Phytol 165: 351–371 mental Panel on Climate Change. Cambridge University Press, Cam-
Ainsworth EA, Ort DR (2010) How do we improve crop production in a bridge, UK, pp 3–29
warming world? Plant Physiol 154: 526–530 Ito S, Ohtake N, Sueyoshi K, Ohyama T (2006) Allocation of photosyn-
Ainsworth EA, Yendrek CR, Skoneczka JA, Long SP (2012) Accelerating thetic products in soybean during early stages of nodule formation. Soil
yield potential in soybean: potential targets for biotechnological im- Sci Plant Nutr 52: 438–443
provement. Plant Cell Environ 35: 38–52 Koester RP, Skoneczka JA, Cary TR, Diers BW, Ainsworth EA (2014)
Alexandratos N, Bruinsma J (2012) World Agriculture towards 2030/2050: Historical gains in soybean (Glycine max Merr.) seed yield are driven by
The 2012 Revision. Food and Agriculture Organization, Rome linear increases in light interception, energy conversion, and partition-
Amthor JS (2010) From sunlight to phytomass: on the potential efficiency of ing efficiencies. J Exp Bot 65: 3311–3321
converting solar radiation to phyto-energy. New Phytol 188: 939–959 Leakey ADB, Ainsworth EA, Bernacchi CJ, Rogers A, Long SP, Ort DR
Amthor JS, Mitchell RJ, Runion GB, Rogers HH, Prior SA, Wood CW (2009) Elevated CO2 effects on plant carbon, nitrogen, and water rela-
(1994) Energy content, construction cost and phytomass accumulation of tions: six important lessons from FACE. J Exp Bot 60: 2859–2876
Glycine max (L.) Merr. and Sorghum bicolor (L.) Moench grown in ele- Lobell DB, Field CB (2007) Global scale climate-crop yield relationships
vated CO2 in the field. New Phytol 128: 443–450 and the impacts of recent warming. Environ Res Lett 2: 014002
Andrews M, Lea PJ, Raven JA, Azevedo RA (2009) Nitrogen use efficiency. Lobell DB, Gourdji SM (2012) The influence of climate change on global
3. Nitrogen fixation: genes and costs. Ann Appl Biol 155: 1–13 crop productivity. Plant Physiol 160: 1686–1697
Balota M, Payne WA, Rooney W, Rosenow D (2008) Gas exchange and Lobell DB, Roberts MJ, Schlenker W, Braun N, Little BB, Rejesus RM,
transpiration ratio in Sorghum. Crop Sci 48: 2361–2371 Hammer GL (2014) Greater sensitivity to drought accompanies maize
Beadle CL, Long SP (1985) Photosynthesis: is it limiting to biomass pro- yield increase in the U.S. Midwest. Science 344: 516–519
duction? Biomass 8: 119–168 Long SP, Ainsworth EA, Leakey ADB, Nӧsberger J, Ort DR (2006a) Food
Bernacchi CJ, Kimball BA, Quarles DR, Long SP, Ort DR (2007) Decreases for thought: lower-than-expected crop yield stimulation with rising CO2
in stomatal conductance of soybean under open-air elevation of [CO2] concentrations. Science 312: 1918–1921
are closely coupled with decreases in ecosystem evapotranspiration. Long SP, Ort DR (2010) More than taking the heat: crops and global
Plant Physiol 143: 134–144 change. Curr Opin Plant Biol 13: 241–248
Botha FC (2009) Energy yield and cost in a sugarcane biomass system. Proc Long SP, Zhu XG, Naidu SL, Ort DR (2006b) Can improvement in pho-
Aust Soc Sugar Cane Technol 31: 1–10 tosynthesis increase crop yields? Plant Cell Environ 29: 315–330
Cassman KG, Grassini P, van Wart J (2010) Crop yield potential, yield Lopes MS, Reynolds MP, Manes Y, Singh RP, Crossa J, Braun HJ (2012)
trends, and global food security in a changing climate. In D Hillel, C Rosenzweig, Genetic yield gains and changes in associated traits of CIMMYT spring
eds, Handbook of Climate Change and Agroecosystems: Impacts, Adaptation, bread wheat in a “historic” set representing 30 years of breeding. Crop
and Mitigation. Imperial College Press, London, pp 37–51 Sci 52: 1123–1131
Ceotto E, Di Candilo M, Castelli F, Badeck FW, Rizza F, Soave C, Volta A, Lutz W, Samir KC (2010) Dimensions of global population projections:
Villani G, Marletto V (2013) Comparing solar radiation interception what do we know about future population trends and structures? Philos
and use efficiency for the energy crops giant reed (Arundo donax L.) and Trans R Soc Lond B Biol Sci 365: 2779–2791
sweet sorghum (Sorghum bicolor L. Moench). F Crop Res 149: 159–166 Manès Y, Gomez HF, Puhl L, Reynolds M, Braun HJ, Trethowan R (2012)
Cirera X, Masset E (2010) Income distribution trends and future food de- Genetic yield gains of the CIMMYT international semi-arid wheat yield
mand. Philos Trans R Soc Lond B Biol Sci 365: 2821–2834 trials from 1994 to 2010. Crop Sci 52: 1543–1552
Dai A (2011) Drought under global warming: a review. Wiley Interdiscip McKendry P (2002) Energy production from biomass. Part 1. Overview of
Rev Clim Chang 2: 45–65 biomass. Bioresour Technol 83: 37–46
Dohleman FG, Heaton EA, Arundale RA, Long SP (2012) Seasonal dy- Ort DR, Long SP (2014) Limits on yields in the Corn Belt. Science 344: 484–
namics of above- and below-ground biomass and nitrogen partitioning 485
in Miscanthus 3 giganteus and Panicum virgatum across three growing Ort DR, Melis A (2011) Optimizing antenna size to maximize photosyn-
seasons. Glob Change Biol Bioenergy 4: 534–544 thetic efficiency. Plant Physiol 155: 79–85

Plant Physiol. Vol. 168, 2015 391

Slattery and Ort

Parry MAJ, Reynolds M, Salvucci ME, Raines C, Andralojc PJ, Zhu XG, Price expected CO2 stimulation in photosynthesis and productivity for soy-
GD, Condon AG, Furbank RT (2011) Raising yield potential of wheat. II. bean grown in the midwestern United States. Plant Physiol 162: 410–423
Increasing photosynthetic capacity and efficiency. J Exp Bot 62: 453–467 Sadras VO, Lawson C, Montoro A (2012) Photosynthetic traits in Australian
Peng S, Khush GS, Virk P, Tang Q, Zou Y (2008) Progress in ideotype wheat varieties released between 1958 and 2007. Field Crops Res 134: 19–29
breeding to increase rice yield potential. Field Crops Res 108: 32–38 Sayer J, Sunderland T, Ghazoul J, Pfund JL, Sheil D, Meijaard E, Venter
Penning de Vries FWT, Jansen DM, ten Berge HFM, Bakema A (1989) M, Boedhihartono AK, Day M, Garcia C, et al (2013) Ten principles for
Simulation of Ecophysiological Processes of Growth in Several Annual a landscape approach to reconciling agriculture, conservation, and other
Crops. Centre for Agricultural Publishing and Documentation, Wage- competing land uses. Proc Natl Acad Sci USA 110: 8349–8356
ningen, The Netherlands Shearman VJ, Scott RK, Foulkes MJ (2005) Physiological processes asso-
Prasad PVV, Boothe KJ, Allen LH Jr, Thomas JMG (2003) Super-optimal ciated with wheat yield progress in the UK. Crop Sci 45: 175–185
temperatures are detrimental to peanut (Arachis hypogaea L.) reproduc- Sinclair TR, Muchow RC (1999) Radiation use efficiency. Adv Agron 65:
tive processes and yield at both ambient and elevated carbon dioxide. 215–265
Glob Change Biol 9: 1775–1787 Sinha SK, Bhargava SC, Goel A (1982) Energy as the basis of harvest in-
Raines CA (2011) Increasing photosynthetic carbon assimilation in C3 plants to dex. J Agric Sci 99: 237–238
improve crop yield: current and future strategies. Plant Physiol 155: 36–42 Slattery RA, Ainsworth EA, Ort DR (2013) A meta-analysis of responses of

Downloaded from https://academic.oup.com/plphys/article/168/2/383/6113608 by guest on 06 January 2024

Ray DK, Mueller ND, West PC, Foley JA (2013) Yield trends are insuffi- canopy photosynthetic conversion efficiency to environmental factors
cient to double global crop production by 2050. PLoS ONE 8: e66428 reveals major causes of yield gap. J Exp Bot 64: 3723–3733
Reynolds M, Foulkes J, Furbank R, Griffiths S, King J, Murchie E, Parry Tilman D, Balzer C, Hill J, Befort BL (2011) Global food demand and the
M, Slafer G (2012) Achieving yield gains in wheat. Plant Cell Environ sustainable intensification of agriculture. Proc Natl Acad Sci USA 108:
35: 1799–1823 20260–20264
Rincker K, Nelson R, Specht J, Sleper D, Cary T, Cianzio SR, Casteel S, Vessey JK, Walsh KB, Layzell DB (1988) Can a limitation in phloem
Conley S, Chen P, Davis V, et al (2014) Genetic improvement of US supply to nodules account for the inhibitory effect of nitrate on nitro-
soybean in maturity groups II, III, and IV. Crop Sci 54: 1419 genase activity in soybean? Physiol Plant 74: 137–146
Rosegrant M, Agcaoili M (2010) Global Food Demand, Supply, and Price Zhu XG, Long SP, Ort DR (2008) What is the maximum efficiency with
Prospects to 2010. International Food Policy Research Institute, Wash- which photosynthesis can convert solar energy into biomass? Curr Opin
ington, DC Biotechnol 19: 153–159
Ruiz-Vera UM, Siebers M, Gray SB, Drag DW, Rosenthal DM, Kimball Zhu XG, Long SP, Ort DR (2010) Improving photosynthetic efficiency for
BA, Ort DR, Bernacchi CJ (2013) Global warming can negate the greater yield. Annu Rev Plant Biol 61: 235–261

392 Plant Physiol. Vol. 168, 2015