Published in Agron. J. 95:1465-1474 (2003).
© American Society of Agronomy
677 S. Segoe Rd., Madison, WI 53711 USA
CORN
Canopy Structure, Light Interception, and Photosynthesis in Maize
D. W. Stewart*,a,
C. Costab,
L. M. Dwyera,
D. L. Smithc,
R. I. Hamiltona and
B. L. Maa
a Agric. and Agri-Food Can., Eastern Cereal and Oilseed Res. Cent., Ottawa, ON, Canada K1A 0C6
b Univ. of Passo Fundo, Passo Fundo, RS, 99001-970, Brazil
c Dep. of Plant Sci., McGill Univ., Ste-Anne-de-Bellevue, QC, Canada H9X 3V9
* Corresponding author (StewartDW{at}agr.gc.ca).
Received for publication May 13, 2002.
 |
ABSTRACT
|
|---|
The amount and distribution of leaf area and leaf angles in a crop canopy determine how photosynthetically active radiation (PAR) is intercepted and consequently influences canopy photosynthesis and yield. Factors such as plant shape, plant populations, and row width will affect these leaf distributions and can occur in an almost infinite number of different combinations. To supplement experimentation, a mathematical model was developed to use measurements of leaf area and leaf angles in two dimensions (with height and across the row) to calculate PAR interception and canopy photosynthesis. Maize (Zea mays L.) hybrids with phenotypic differences were planted at several plant populations to produce a wide range of two-dimensional leaf area and leaf angle patterns. The extreme phenotypes, leafy and reduced stature, were included to vary plant height and number of leaves above the ear. Measurements of average PAR at various levels were made in seven different canopies and compared with calculations from the model (R2 of 0.68 and 0.92 for two sets of data). As well, measurements of PAR at 20-cm increments on transects perpendicular to the row were made in three canopy types at three levels and compared with theoretical calculations (R2 = 0.74). A simple numerical experiment was run to demonstrate the utility of the model where daily canopy photosynthesis was calculated for two row widths and seven plant types. One result was that depending on row widths, plants with very upright leaves can have both the smallest and largest daily canopy photosynthesis.
Abbreviations: PAR, photosynthetically active radiation • RMSE, root mean square error
|
INTRODUCTION
|
|---|
HOW PAR IS INTERCEPTED by the plant canopy is a crucial factor in determining canopy photosynthesis and crop yield. Considerable effort has been expended on theoretical models of light interception (de Wit, 1965; Anderson, 1966; Norman and Jarvis, 1974; Goudriaan, 1977; Norman, 1980; Campbell, 1986). Most of this effort considered variation in canopy architecture in only the vertical dimension (height). The ability to characterize the two-dimensional aspects of leaf area and leaf angle distribution (with height and across rows) in maize is important as considerable variation has been observed in leaf area distribution across rows (Stewart and Dwyer, 1993). An important aspect of this problem is to consider how a plant rotates its leaves into three-dimensional space (Espana et al., 1999). Theoretical aspects of two-dimensional light interception by crops have been derived by Allen (1974), Norman and Welles (1983), Sinoquet (1989), and Sinoquet and Bonhomme (1992), among others. However, to date, measured two-dimensional leaf area and leaf angle data have not been incorporated into a light interception–canopy photosynthesis model.
Two-dimensional aspects of light interception are particularly important in mid- to short-season production areas where leaf area index is rarely in excess of that required to maximize canopy light interception. Development of new phenotypes for these production areas, including leafy and leafy–reduced stature (Dwyer et al., 1995b; Begna et al., 1997; Modarres et al., 1997), has required reassessment of optimum plant populations and planting patterns to maximize production. The objectives of this study were to develop methods to quantify two-dimensional leaf area distribution and to use this distribution to calculate light interception and canopy photosynthesis. This methodology was used to characterize the two-dimensional distribution of leaf area of maize hybrids with contrasting architecture and to compare calculations of light penetration into these canopies with measurements. We also used the theory to calculate PAR flux densities on leaf surfaces, which were in turn used within a photosynthesis function to calculate canopy photosynthesis. Canopy photosynthesis was used to evaluate how effectively different plant shapes and row widths used incoming PAR.
|
MATERIALS AND METHODS
|
|---|
Theory
The following one-dimensional equation has been used to describe PAR penetration into a canopy with randomly dispersed leaves:
 | [1] |
where I is the average PAR flux density at a given height in the canopy; I0 is the PAR flux density at the top of the canopy; 
is the angle of the sun's elevation; LA is leaf area index accumulated from the top of the canopy to the given height; and PL is the mean projection of a leaf element in the direction of the sun's rays (Duncan et al., 1967; Stewart and Lemon, 1969; Lemeur and Blad, 1974; Norman and Welles, 1983). The angle between the direction of the sun's rays and the normal to the leaf (
) is related to the leaf angle (ß) (i.e., the angle between the leaf plane and horizontal), angle of the sun's elevation (), and the angle between the azimuths of the sun's rays and the leaf (
) by:
 | [2] |
where
 | [3] |
and
 | [4] |
The mean leaf projection (PL) is obtained by integrating sin() around the azimuth, which results in:
 | [5] |
and
 | [6] |
where
 | [7] |
These equations have been derived by de Wit (1965), Duncan et al. (1967), and Lemeur and Blad (1974).
For two dimensions (height and the direction perpendicular to the row), the canopy is divided into increments (i,j) with respect to vertical and horizontal distances. Using the above procedures, leaf area and leaf angle distributions are measured for each increment (i,j) and stored in matrices. Then PAR flux density intercepted by an increment (i,j) is calculated from:
 | [8] |
where Ii,j is the PAR flux density at the bottom of the increment, Ii-1,j is the PAR flux density at the top of the increment, LAi,j is the leaf area per horizontal area of the increment, and PLi,j is the mean leaf projection. Note that both leaf area and the mean projection are stored in two-dimensional arrays. The mean projection used leaf angles, which were also stored in two-dimensional arrays.
The path of the sun's rays in the canopy had to be followed to keep track of increments. This involved the following relationship between row direction and the angle of the sun's rays, which are illustrated in Fig. 1
. From Fig 1a:
 | [9] |
and from Fig. 1b:
 | [10] |
where 
is the angle between the sun's rays and the row. As the ray penetrates into the canopy, 
x is the horizontal distance the ray travels perpendicular to the row, and y is the vertical distance. Combining Eq. [9] and [10] by eliminating BB', which is the distance the ray travels in the horizontal plane, resulted in:
 | [11] |
Starting at the top of the canopy, the distance down is in steps of 
y so that y is always an integer times y. Each y corresponded to the ith increments exactly. The x's did not correspond to the jth increments exactly as x is not an integer times x except when sin() equaled tan(). Simple linear interpolation was used to calculate values of leaf area and leaf angle for each x across each row so that PAR flux densities could be calculated for each y and x.
The fraction of leaf area (unit leaf area per unit soil area) exposed to direct PAR was expressed by Duncan et al. (1967) as:
 | [12] |
where AHi,j and AHi-1,j are fractions of horizontal areas exposed to direct PAR at the top and bottom of the i,jth increment. These horizontal fractions can be calculated from:
 | [13] |
A portion of PAR was reflected or transmitted by leaves. The average PAR flux density falling on a leaf in direct sunshine (DA) was expressed as:
 | [14] |
Transmitted and reflected PAR are DA
and DA
, respectively, where and are leaf transmissivity and reflectivity, respectively. When PAR is transmitted or reflected, it is assumed to scatter diffusely or equally in all directions. The proportion of PAR reflected downward or upward or transmitted downward or upward (E) is calculated by integrating the hemisphere surrounding the plane of the leaf element (Stewart and Lemon, 1969), resulting in:
 | [15] |
The total amount of PAR scattered upward (Sc
) or downwards (Sc
) in the i,jth increment was given by:
 | [16] |
and
 | [17] |
The PAR scattered downward was added to amounts of diffuse PAR coming from the sky. For this purpose, the sky was assumed to be uniformly bright and divided into nine concentric rings, each with an arc of 10°. For example, the first ring would start at the horizon and end at 10° above the horizon, the next would start at 10° and end at 20°, etc. The relative area and thus the relative amount of radiation from each segment of each concentric ring was calculated by integrating surface areas on a hemisphere. Ten percent of PAR arriving downward at the top of the canopy was assumed to come from the sky for the conditions of Exp. A described here, that is, relatively cloud-free sky conditions. For Exp. B, the percentage of sky radiation was measured. Then PAR from each concentric ring was treated the same as direct PAR from the sun. All diffuse PAR was summed (sky PAR from all segments and PAR scattered by leaves) for each vertical and horizontal increment in the canopy.
The final objective of the light interception model was to calculate canopy photosynthesis. This was done by using PAR flux densities at leaf surfaces in a leaf photosynthesis model and integrating photosynthesis per unit leaf area over the canopy (Stewart and Lemon, 1969). Because photosynthesis has a nonlinear response to PAR, average PAR flux densities were considered inadequate (Norman, 1980). The above theory calculated the PAR flux densities at each level in the canopy with the proportion of sunlit and shaded area as well as amounts of diffuse PAR. Similar theory was used to calculate PAR flux densities at leaf surfaces. In Eq. [1], I0 represented radiation from either the sun or from a concentric sky ring. I0DIR was defined as the flux density of radiation coming directly from the sun and falling on a horizontal surface. The flux density of this radiation on a plane perpendicular to the sun's rays (IDIR) was
 | [18] |
and the corresponding flux density on a sunlit leaf (ILDIR) in the canopy was:
 | [19] |
ILDIR was calculated for each 10° leaf position as a leaf was turned through 360° around the azimuth calculating from Eq. [2], [3], and [4], assuming that leaf angle distributions in two dimensions are known. Therefore, at each level in the canopy, we calculated the relative amount of leaf area in sun and shade from Eq. [12] and [13]. For the leaf area in the sun, we calculated a range of flux densities from Eq. [19] and added diffuse radiation. This information was stored in 20 PAR flux density classes for each horizontal increment in the canopy. That is, we summed the relative amount of leaf area that received flux densities in each class. The first class would have a flux density of a leaf oriented perpendicular to the direction of the sun's rays in full sunlight. The 20th class would have a flux density 1/20 of the first class. Most of the shaded leaf area fell into this class. The PAR values in each class were represented by ILik and the leaf area by ALik where i was the horizontal canopy increment and k the PAR flux density class.
Canopy photosynthesis can be calculated at any time during the day using astronomical equations to determine sun angle if diffuse and direct components of PAR are known. Photosynthesis in each PAR class (Pik) was calculated from:
 | [20] |
where 
was an empirical coefficient, D was leaf respiration, and P2000 was the photosynthesis rate at a PAR flux density of 2000 µmol m-2 s-1 (Dwyer and Stewart, 1986). Canopy photosynthesis (PC) was expressed as:
 | [21] |
where LAI was the leaf area index of the canopy.
Measurements
There were two types of measurements: plant measurements to characterize canopies and PAR measurements to compare with theoretical calculations. Plants were selected from plots from field experiments at two locations. In Exp. A, four maize hybrids were planted in six rows in a randomized split-plot design with three replications at Ste-Anne-de-Bellevue, QC (47° N, 71° W), in 1995. The main-plot treatment consisted of two N levels (170 and 255 kg N ha-1). Subplot treatments were the four hybrids, representing conventional (C) (‘Pioneer 3979’), leafy (Lfy), reduced stature (rd1), and leafy–reduced stature (Lfy-rd1) phenotypes (Modarres et al., 1997). Each hybrid was grown in 0.76-m rows at planting densities that increased as the size of the hybrid decreased: C at 75000 plants ha-1, Lfy at 55000, rd1 at 100000, and Lfy-rd1 at 90000. Note that in this study, these treatments were used only to provide plant stands with a wide range of canopy types. We were not studying effects of plant populations by analysis of variance. In Exp. B, performed at Ottawa, ON (45° N, 75°W), in 1998, three hybrids at 60000 plants ha-1 were planted in a randomized complete block design with three replications in 12-row plots with 0.76-m row widths. These hybrids were ‘Yuyu5’, a hybrid from China characterized by very upright leaves; a conventional hybrid, ‘Pioneer 3751’; and a commercial leafy hybrid, ‘Mycogen TMF 94’. Fertilizer P and K in both experiments were applied in accordance with recommendations from soil analysis. In Exp. B, 200 kg N ha-1 was incorporated before planting.
In both experiments, measurements of two-dimensional leaf area distribution were made near silking following the methods of Stewart and Dwyer (1993). Three to five plants selected at random from each plot were placed against a vertical grid. The following measurements were made on each leaf per plant: (i) the height of the leaf ligule, (ii) the angle between the leaf and the horizontal plane at the leaf ligule (), (iii) the coordinates of the maximum height of the leaf (XM, YM), and (iv) the coordinates of the leaf tip (XL, YL).
Individual leaf measurements also included leaf width at the ligule and every 10 cm along the leaf and total leaf length. Leaf width was related to length along the leaf by the following second-order polynomial as derived by Stewart and Dwyer (1999):
 | [22] |
where w was leaf width, WM was maximum leaf width, z was distance from the ligule divided by the total leaf length, and c1 and c2 were empirical coefficients. Plant height and stem diameter at the top, middle, and bottom of the plant were also measured. The curvature of the leaf (the curved line passing through the origin, the maximum coordinates, and the coordinates of the leaf tip) was represented by a second-order function
 | [23] |
where x is the horizontal distance from the origin or ligule, y is the vertical distance, and A, B, C, D, E, and G are empirical leaf curvature coefficients. Following procedures described in Stewart and Dwyer (1993), leaf area and leaf angle distributions were calculated with height and across the horizontal plane of the grid. To calculate leaf area across the row, the plane was rotated mathematically from 0 to 90° in steps of 10°. At each position, the area of each finite increment was projected against the plane of the row and averaged over the 10 positions. Averaging was weighted because of a tendency for leaves to grow into the rows rather than along rows. From measurements by Girardin (1992) and our own observations, we weighted plants growing perpendicular to the row by a factor of 2 compared with a factor of 1 for plants growing along the row. Weighting factors for the other positions were determined by linearly interpolating between 1 and 2 according to the degree of turn.
The diameter of each stem was measured at the plant tip, at cob height, and at the ground surface. Diameters at all heights were calculated by linearly interpolating between the measurements. Stems were assumed to be cylinders, and the cross-sectional area of segments were calculated for each height increment and added to the leaf area for each increment. This area was assumed to fall into the 80 to 90° leaf angle class. This meant that some of the area of the tassel was accounted for but not all of it.
Light interception was measured in both experiments in the same plots from which plants were sampled. In Exp. A, PAR interception (µmol m-2 s-1) was measured on 13 Aug. 1995 (near silking) when sun angles were high, with cloud-free sky conditions, and when most of the PAR came directly from the sun. Measurements were made above the crop and at four canopy heights (ground level and 50, 100, and 150 cm) using 1-m line quantum sensors (LI-2000, LI-COR, Lincoln, NE) leveled and placed diagonally between the two central rows (the tip of the probe in one row and the end of the probe in the other row) of each plot at the four heights in each canopy. These 1-m sensors measured an average value of PAR at each level. Measurements were made, simultaneously, at each of the five levels in a plot. This was important for comparisons with the theory as the measured PAR value at the top of the canopy was used in the theory to calculate values of PAR in the canopy. The PAR interception measurements were obtained in each treatment a minimum of three times during the day, and the exact time of measurement was recorded.
In Exp. B, PAR interception was measured across rows of three hybrids in three plots at three levels in the canopy (ground, ear height, and midway between the ear and the top of the canopy) with three replications using an AccuPAR probe (Decagon Devices, Pullman, WA). This 0.8-m probe was centered on a row and placed perpendicular to the row to measure PAR interception at each centimeter along the probe. These 1-cm measurements were averaged over 5 cm. In Exp. B, only one bar was available, so the bar was moved from level to level to obtain the three sets of measurements in each canopy. Three plants were sampled from each plot, and leaf area and leaf angles were measured as described above. Again, PAR measurements were made on clear days to minimize diffuse radiation from sky and clouds. Diffuse and direct PAR above the canopy were measured with each set of in-canopy measurements using the AccuPAR system.
Interception of PAR was calculated using the equations in the theoretical section and leaf measurements. We calculated average PAR values at each level where measurements were made as well as PAR values as they varied across the row. This was done in both Exp. A and Exp. B, but only in Exp. B did we have measurements to compare with theoretical calculations across the row. For each set of calculations, we measured the angle between row direction and solar north and the amount of direct and diffuse PAR. Using this angle, time of day, and day number, the sun elevation angle and the angle between the sun's rays and the row direction were calculated using astronomical equations (List, 1958). Knowing the leaf area distribution across the row and with height in the canopy, PAR interception across the row and with height in the canopy was calculated based on the above theory.
Average values of PAR at the required levels in the various canopies were calculated to compare with measurements in Exp. A. A root mean square error (RMSE) was calculated for each set of measurements and was defined as:
 | [24] |
where IM = measured PAR values, IC = calculated PAR values, and n = the number of measurements. Comparisons were also evaluated using simple correlation and linear regression. Linear regression was used only to study bias in the comparison. For Exp. B, the theory was used to calculate values averaged over each 5- and 20-cm increment across rows at the appropriate levels in the canopies to compare with the AccuPAR measurements. Again RMSE, correlation, and linear regression were used to evaluate the comparisons.
Numerical Experiments
The model was then used in a simple numerical experiment where canopy photosynthesis was calculated every hour from sunrise to sunset to obtain daily totals for a clear day at tasseling for all hybrids at the plant populations and row width (78 cm) used in this study. Diffuse light from the sky was assumed to be 10% of the total amount of PAR coming from the sun and sky. Daily canopy photosynthesis was then calculated for the three plant types of Exp. B for row widths of 76 and 38 cm and a plant population of 60000 plants ha-1 using Eq. [20] and [21]. Values for , P2000, and D were 0.0023 mg CO2 µmol-1, 1.52 mg CO2 m-2 s-1, and 0.11 mg CO2 m-2 s-1 from measurements of Dwyer and Stewart (1986). This numerical experiment was done to demonstrate how this model could be used. Clearly, there are an unlimited number of different what if scenarios. In this example, we are not calculating actual canopy photosynthesis. We are attempting to isolate the effect of architecture on canopy photosynthesis by changing the row width for contrasting plant shapes.
|
RESULTS AND DISCUSSION
|
|---|
The hybrids used in this study provided a wide range of plant types, with total height, total number of leaves, the number of leaves above the ear, and leaf area index varying from 0.79 to 2.76 m, 9.4 to 14.7, 5.7 to 9.6, and 1.5 to 4.7, respectively (Table 1). While plots were used with two levels of N in Exp. A, no significant differences were found between the two N levels (data not shown). Therefore, data in Table 1 were averaged over the two N treatments. These plants at their respective populations display large variations in two-dimensional leaf area with height and along transects perpendicular to the row. This is illustrated in Fig. 2
where only three contrasting hybrids are shown. The leaf angles were relatively similar among hybrids, except for Yuyu5, which had very upright leaves (Fig. 3)
. While upright leaves may in certain circumstances improve the efficiency of PAR utilization, they have the negative quality of keeping leaves close to the stalk in maize and, therefore, allowing more radiation to penetrate to the soil surface. Upright leaves were largely responsible for the very large gradients in leaf area distribution across the row for Yuyu5 (Fig. 2a). Therefore, to accurately evaluate the importance of leaf angles in maize, the two-dimensional shape of the plant had to be adequately simulated in a mathematical model. Both leaf area and leaf angles have to be incorporated in two dimensions to adequately describe the plant phenotype.
View this table:
[in this window]
[in a new window]
|
Table 1. Values of plant height (Ht), total number of leaves (TNL), ear leaf number (EL) (assuming the top leaf is 1), and leaf area index (LAI) for the seven hybrids used in this study.
|
|
View larger version (70K):
[in this window]
[in a new window]
|
Fig. 2. Leaf area densities (m2 m-3) across the row and with crop height for three hybrids: (a) the Chinese hybrid Yuyu5, (b) Pioneer 3861, and (c) Micogen TMF 94.
|
|
View larger version (29K):
[in this window]
[in a new window]
|
Fig. 3. Average cumulative frequency of leaf angle distributions for the following seven hybrids: (a) the Chinese hybrid Yuyu5, ‘Pioneer 3791’, and Micogen TMF 94; (b) Rd (a reduced-stature hybrid, Rd x Lfy (a hybrid from crossing reduced-stature inbred with a leafy inbred), ‘Pioneer 3879’, and Lfy (a leafy hybrid).
|
|
In Exp. A, a significant correlation (P 
0.05) was obtained between measured and calculated PAR values (R2 = 0.68, RMSE = 239.1 µmol m-2 s-1) (Fig. 4)
. While scatter is evident, the slope of the linear regression of measured vs. calculated values was not significantly different from 1, and the y intercept was not significantly different from 0, indicating there was no bias in the model (data not shown).
View larger version (23K):
[in this window]
[in a new window]
|
Fig. 4. Average values of measured vs. calculated photosynthetically active radiation (PAR) values (µmol m-2 s-1) at ground level and 50, 100, and 150 cm in canopies for the four hybrids of Exp. A. The R2 was equal to 0.68 with a root mean square error of 239.1 µmol m-2 s-1.
|
|
There was more scatter in the comparison of observed and calculated PAR values across the row in Exp. B (R2 = 0.61, RMSE = 432.0 µmol m-2 s-1) (Fig. 5)
. This scatter reflected the increased difficulty of obtaining representative measurements along transects perpendicular to the row compared with average values although this R value was significant (P 0.05). When values were averaged over 20-cm intervals, the correlation improved (R2 = 0.74, RMSE = 174.0 µmol m-2 s-1) (Fig. 5). This improved to an R2 of 0.92 and a RMSE of 106.4 µmol m-2 s-1 when all values for each level in the canopy were averaged (Fig. 5). Figure 5c indicates some bias as measured PAR was higher than calculated PAR, particularly for low PAR values. We hypothesize that random gaps in the row resulted in greater light penetration, particularly at lower sun angles, than were calculated for the average plant population density.
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 5. Measured vs. calculated photosynthetically active radiation (PAR) values at three levels (ground surface, cob height, and halfway between cob height and canopy top) and three times (1000, 1100, and 1200 h EST) for three hybrids in Exp. B. Values are (a) averages of every 5 cm across the row, (b) every 20 cm across the row, and (c) all values averaged per level. Root mean square errors were 432, 174, and 106 µmol m-2 s-1, respectively.
|
|
The theoretical values of PAR across each row and level were realistic (Fig. 6, 7, and 8)
. The sun-row angles were approximately 42, 30, and 5° for the 1000, 1100, and 1200 h EST (Eastern Standard Time; i.e., the standard time for our time zone) while the sun elevation angles were approximately 37, 44, and 52°. Therefore, the sun elevation angles increased while row angles decreased over the three times. The theoretical calculations clearly showed that the shadow of a row was apparent at lower sun elevation angles and larger row angles but disappeared at 1200 h when the plane of the sun's rays were almost parallel to the plane of the row and the sun was at it's highest elevation. Also evident were the larger amounts of PAR penetrating down the rows for the Chinese hybrid (Yuyu5) with more erect leaves (Fig. 6).
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 6. Measured and calculated photosynthetically active radiation (PAR) values across rows for three times [(a) 1000, (b) 1100, and (c) 1200 h EST] and three heights for Yuyu5 in Exp. B. Model calculations are for every 5 cm across the row. Measurements are averaged over 20 cm. The three heights correspond to the ground surface, cob height, and halfway between the cob height and the top of the canopy.
|
|
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 7. Measured and calculated photosynthetically active radiation (PAR) values across rows for three times [(a) 1000, (b) 1100, and (c) 1200 h EST] and three heights for Pioneer 3861 in Exp. B. Model calculations are for every 5 cm across the row. Measurements are averaged over 20 cm. The three heights correspond to the ground surface, cob height, and halfway between the cob height and the top of the canopy.
|
|
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 8. Measured and calculated photosynthetically active radiation (PAR) values across rows for three times [(a) 1000, (b) 1100, and (c) 1200 h EST] and three heights for Mycogen TMF 94 in Exp. B. Model calculations are for every 5 cm across the row. Measurements are averaged over 20 cm. The three heights correspond to the ground surface, cob height, and halfway between the cob height and the top of the canopy.
|
|
Calculations of daily photosynthate production showed a typical exponential response to leaf area (Fig. 9)
. It was interesting that the upright variety (Yuyu5) falls below the average response curve. Therefore, the negative clumping of leaf area around the stalks decreased the canopy efficiency of Yuyu5. As an example of how this model may be used, the following simple numerical experiment was conducted. Row width was reduced by one-half in the model but with the same plant populations. Daily photosynthate production was calculated assuming that the shape and leaf area of individual plants did not change with row width and using a constant plant population. The daily photosynthate production of the Yuyu5 canopy increased by 17.2% compared with a 6.4% increase for ‘Pioneer 3861’ and a 3.76% increase for Mycogen TMF 94 when row widths were reduced by one-half (Table 2). Decreasing the row width decreased the leaf area gradients across the row, and this effect was greatest with Yuyu5 (Fig. 10)
.
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 9. Canopy photosynthate production per day as a function of leaf area index {Y = 47.42[1 - exp(-0.8148X)], R2 = 0.73, SEE = 3.735 g m-2 d-1, where SEE = standard error of the estimate}.
|
|
View this table:
[in this window]
[in a new window]
|
Table 2. Canopy photosynthate production for a 76-cm row (P1) and a 38-cm row (P2) at measured values of leaf area index (LAI).
|
|
View larger version (20K):
[in this window]
[in a new window]
|
Fig. 10. Simulated leaf area index variation across the row for (a) 76-cm row width and (b) 38-cm row width.
|
|
The photosynthesis values were calculations from the model and were not real values of net photosynthesis. Similar calculations of gross photosynthesis from one-dimensional models have been presented by de Wit (1965), Duncan (1971), and Goudriaan and van Laar (1978). There were many simplifications, such as a simple constant to represent leaf respiration (D). We don't consider stalk and ear respiration. P2000 changes with both age, leaf number in the canopy, and hybrid (Dwyer et al., 1989, 1995a; Stewart et al., 1997). However, by keeping , P2000, and D constant with leaf number and hybrid, we showed how only architectural changes in the canopy would affect canopy photosynthesis. Duncan (1971) simply varied leaf angle to calculate plant shape effects on canopy photosynthesis. In this study, we captured a more realistic shape of hybrids and showed two aspects of leaf angles. One was the effect of PAR flux densities on leaves and therefore on photosynthesis. The other was the clumping effect leaf angles had on the distribution of leaf area across the row and the consequent decrease of light interception. Plant breeders should consider row widths in developing ideal plant shapes. There are some reservations. Leaves are treated as flat ribbons. This is compensated to some extent by how leaves are positioned in leaf photosynthesis chambers, particularly the type where the whole cross section of the leaf is contained in the chamber. In this situation, it is impossible to totally flatten the leaf, and PAR flux densities fall on the undulating surfaces.
These results suggest that the shape of the maize plant can be accurately simulated using two-dimensional distributions of both leaf area and leaf angles. The model may have applications in quantifying the effect of canopy structure and planting patterns on crop photosynthesis and, when integrated over the growing season, on crop yield. Future research will be directed toward using this model to design plant canopies for optimum light interception. Research is being conducted on a wide range of canopy types. Of particular interest is identifying the position in the canopy where light is intercepted and where carbohydrates are being produced and the implications of this spatial variation for production of grain yield. Results reported here indicated that the proposed model of leaf area distribution and PAR interception will be useful for quantifying the effect of canopy structure and planting patterns on crop photosynthesis.
|
ACKNOWLEDGMENTS
|
|---|
The authors gratefully acknowledge the excellent technical assistance of B. Wilson, L. Ranacher, L. Evenson, D. Balchin, D. Meridith, and P. Neaves.
|
NOTES
|
|---|
Eastern Cereal and Oilseed Research Centre Contribution no. 03-256.
|
REFERENCES
|
|---|
- Allen, L.H., Jr. 1974. Model of light penetration into a wide-row crop. Agron. J. 66:41–47.[Abstract/Free Full Text]
- Anderson, M.C. 1966. Stand structure and light penetration: II. A theoretical analysis. J. Appl. Ecol. 3:41–54.
- Begna, S., R.I. Hamilton, L.M. Dwyer, D.W. Stewart, and D.L. Smith. 1997. Effects of population density on the vegetative growth of leafy reduced-stature maize in short season areas. J. Agron. Crop Sci. 182:4–55.
- Campbell, G.S. 1986. Extinction coefficients for radiation in plant canopies calculated using an ellipsoidal inclination angle distribution. Agric. For. Meteorol. 36:317–321.
- de Wit, C.T. 1965. Photosynthesis of leaf canopies. Agric. Res. Rep. 663. Inst. for Biol. and Chem. Res. on Field Crops and Herbage, Wageningen, the Netherlands.
- Duncan, W.G. 1971. Leaf angles, leaf area, and canopy photosynthesis. Crop Sci. 11:482–485.[Abstract/Free Full Text]
- Duncan, W.G., R.S. Loomis, W.A. Williams, and R. Hanau. 1967. A model for simulating photosynthesis in plant canopies. Hildgardia 38:181–205.
- Dwyer, L.M., A.M. Anderson, D.W. Stewart, B.L. Ma, and M. Tollenaar. 1995a. Changes in maize hybrid photosynthetic response to leaf nitrogen, from preanthesis to grain fill. Agron. J. 87:1221–1225.[Abstract/Free Full Text]
- Dwyer, L.M., C.J. Andrews, D.W. Stewart, B.L. Ma, and J.-A. Dugas. 1995b. Analysis of carbohydrate levels in field-grown leafy and normal maize genotypes. Crop Sci. 35:1020–1027.[Abstract/Free Full Text]
- Dwyer, L.M., and D.W. Stewart. 1986. Effects of leaf age and position on net photosynthetic rates in maize (Zea mays L.) Agric. For. Meteorol. 37:29–46.
- Dwyer, L.M., D.W. Stewart, D. Balchin, L. Houwing, C.J. Marur, and R.I. Hamilton. 1989. Photosynthetic rates of six maize cultivars during development. Agron. J. 81:597–602.[Abstract/Free Full Text]
- Espana, M.L., F. Baret, F. Aries, M. Chelle, B. Andrieu, and L. Prévot. 1999. Modelling maize canopy 3D architecture application to reflectance simulation. Ecol. Modell. 122:24–43.
- Girardin, Ph. 1992. Leaf azimuth in maize canopies. Eur. J. Agron. 1:91–97.
- Goudriaan, J. 1977. Crop micrometeorology: A simulation study. Simulation Monogr. Pudoc, Wageningen, the Netherlands.
- Goudriaan, J., and H.H. van Laar. 1978. Calculations of daily totals of the gross CO2 assimilation of leaf canopies. Neth. J. Agric. Sci. 26:373–382.
- Lemeur, R., and B.L. Blad. 1974. A critical review of light models for estimating the shortwave radiation regime of plant canopies. Agric. For. Meteorol. 14:255–286.
- List, R.J. 1958. Smithsonian meteorological tables. Smithsonian Inst., Washington, DC.
- Modarres, A.M., R.I. Hamilton, L.M. Dwyer, D.W. Stewart, D.E. Mather, M. Dijak, and D.L. Smith. 1997. Leafy reduced-stature maize for short-season environments: Morphological aspects of inbred lines. Euphytica 96:301–309.
- Norman, J.M. 1980. Interfacing leaf and canopy light interception models. p. 49–84. In J.D. Hesketh and J.W. Jones (ed.) Predicting photosynthesis for ecosystem models. CRC Press, Boca Raton, FL.
- Norman, J.M., and P.G. Jarvis. 1974. Photosynthesis in sitka spruce (Picea sitchensis (Bong.) Carr.): III. Measurements of canopy structure and interception of radiation. J. Appl. Ecol. 11:375–398.
- Norman, J.M., and J.M. Welles. 1983. Radiative transfer in an array of canopies. Agron. J. 75:481–488.[Abstract/Free Full Text]
- Sinoquet, H. 1989. Modélisation de l'interception des rayonnements solaires dans une culture en rangs: I. Aspects théoriques. Agronomie 9:125–135.
- Sinoquet, H., and R. Bonhomme. 1992. Modeling radiative transfer in mixed and row intercropping systems. Agric. For. Meteorol. 62:219–240.
- Stewart, D.W., and L.M. Dwyer. 1993. Mathematical characterization of maize canopies. Agric. For. Meteorol. 66:247–265.
- Stewart, D.W., and L.M. Dwyer. 1999. Mathematical characterization of leaf shape and area in maize hybrids. Crop Sci. 39:422–427.[Abstract/Free Full Text]
- Stewart, D.W., L.M. Dwyer, C.J. Andrews, and J.-A. Dugas. 1997. Modeling carbohydrate production, storage, and export in leafy and normal maize (Zea mays L.). Crop Sci. 37:1228–1236.[Abstract/Free Full Text]
- Stewart, D.W., and E.R. Lemon. 1969. A simulation of net photosynthesis of field corn. U.S. Army ECOM Tech. Rep. 2–681–6. U.S. Army Electronic Command, Fort Huachuca, AZ.
This article has been cited by other articles:
|
|
|
|
 
K. D. Subedi, B. L. Ma, and D. L. Smith
Response of a Leafy and Non-Leafy Maize Hybrid to Population Densities and Fertilizer Nitrogen Levels
Crop Sci.,
July 25, 2006;
46(5):
1860 - 1869.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
B. L. Ma, K. D. Subedi, D. W. Stewart, and L. M. Dwyer
Dry Matter Accumulation and Silage Moisture Changes after Silking in Leafy and Dual-Purpose Corn Hybrids
Agron. J.,
June 5, 2006;
98(4):
922 - 929.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
K. D. Subedi and B. L. Ma
Ear Position, Leaf Area, and Contribution of Individual Leaves to Grain Yield in Conventional and Leafy Maize Hybrids
Crop Sci.,
September 23, 2005;
45(6):
2246 - 2257.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
