Agricultural and Forest Meteorology 139 (2006) 7483 www.elsevier.

com/locate/agrformet

Radiation interception and use by maize/peanut intercrop canopy

M.A. Awal a,*, H. Koshi a, T. Ikeda b
a

Graduate School of Science and Technology, Niigata University, Niigata 950-2181, Japan b Faculty of Agriculture, Niigata University, Niigata 950-2181, Japan

Received 5 November 2004; received in revised form 10 May 2006; accepted 1 June 2006

Abstract The efcient use of solar radiation is one of the major criteria for obtaining a yield advantage through intercropping. Although various combinations of crops have been reported for intercropping systems, the maize/peanut association has yet to be analysed. This report presents the radiation-use efciency (e) results of a maize/peanut intercrop study. The experiment constituted three treatments: sole crops of maize and peanut, and a maize/peanut intercrop. The canopy light extinction coefcient (k) of peanut was reduced while intercropped with maize. The mean e of intercropped peanut (2.13 g(DW) MJ1) was 79% higher than that of peanut stands alone. The e of combined intercropped stands (3.03 g(DW) MJ1) was more than two-fold that of sole peanut, but slightly lower than that of maize stands alone (3.27 g(DW) MJ1). The harvest index (HI) of intercropped peanut was about 13% lower than that of peanut grown alone, but produced 46% of the pods of the latter (299 g m2), the parameter that represents the true output of this intercropping system. These results suggest that a maize/peanut intercropping would help to increase production through the efcient utilisation of solar energy. A simple model was developed to isolate the daily radiation interception of each of the canopies of the intercrop partners in separate strata, taking into account the alteration of the k of the understorey. The model may also be applicable in agroforestry systems. # 2006 Elsevier B.V. All rights reserved.
Keywords: Light extinction coefcient; Light partitioning; Spatial transmission; Subordinate canopy

1. Introduction In the modernisation of crop production, most methods used by growers to achieve better yields have largely been explored. Most of these methods involve increasing the efciency of the utilisation of natural resources such as the water and nutrients in the soil, land area, solar radiation, and atmospheric CO2. Many of

* Corresponding author at: Graduate School of Bioagricultural Sciences, Nagoya University, Furo-cho, Chikusa-ku, Nagoya 4648601, Japan. Tel.: +81 52 789 4061; fax: +81 52 789 5038. E-mail address: awalma7@yahoo.com (M.A. Awal). 0168-1923/$ see front matter # 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.agrformet.2006.06.001

these natural resources are becoming more limited, which constitutes a threat. However, the intensity of solar radiation will remain relatively constant, and represents a resource that could be used more efciently for crop production. The abundant radiation available over the tropics and subtopics presents a great opportunity to increase its use for better crop production. The amount of cultivable land is gradually decreasing, mainly because of rapid urbanisation and industrialisation due to the global population explosion. The limited land areas are facing pressure to meet basic demands, especially for food, bre, and oil. Most growers own very small plots of land, especially in the developing countries of Asia and Africa. These growers

M.A. Awal et al. / Agricultural and Forest Meteorology 139 (2006) 7483

75

require not only increased production, but also the ability to grow multiple crops in small areas. Intercropping is one sustainable idea that can greatly increase the use of solar radiation. The additional solar energy used by the intercrop canopy leads to improved crop production, and thus greater economic yield. Over the past three decades, there has been much study on the uses of radiation in intercropping and alley cropping systems for a wide range of crop combinations (Ofori and Stern, 1987; Corlett et al., 1992; Keating and Carberry, 1993; Black and Ong, 2000, and references therein). However, the association of maize and peanut had not yet been analysed. In intercropping systems, the combination of partner crops depends mainly on the crop geometry, but also on the growth habit, life span, and management practices of the crops; climatic conditions; local food habits; growers demands; and nally the benets, whether in terms of crop yield, economics, stability, or sustainability (Connolly et al., 2001). In most cases, dominant and subordinate partners constitute the intercrop mixture. Generally, the former partner is highly dominant over the latter, and is a monocotyledonous cereal due to the erectophilous architectures and isobilateral leaf characteristics of these species. The erect canopy absorbs incidental radiation very effectively through both leaf surfaces, while allowing a sufcient amount of radiation to reach the subordinate crop. In contrast, the subordinate species is selected from those with planophile canopies, which trap the maximum amount of radiation transmitted from the dominant crop. The canopy geometries of the dominant and subordinate stands form different types of growing mixtures: two (or more) separate strata, separate below/ mixed above or vice versa, or completely intermingled. In intercrops with separate strata, the dominant canopy grows independently of the subordinate species with almost no changes in its canopy architecture or radiation characteristics. In contrast, the canopy geometry of the subordinate species is likely inuenced to a great extent by the shading offered by the dominant canopy, but information on the underlying concept is still lacking. To compute their individual e values of the species, it is essential to determine the proportions of radiation intercepted by the partners on a daily basis. Several models have been reported in this regard (Marshall and Willey, 1983; Rimmington, 1984, 1985; Wallace et al., 1991; Sinoquet and Bonhomme, 1992; Sinoquet et al., 2000). However, the application of those models is very complex, and a large number of input variables and complicated mathematics would be required. Moreover, such models for intercrops with separate strata would be

verbose in practice. Most of the models are based on Beers law, and interestingly, this law is perfectly appropriate for crop mixtures in which the canopies form separate vertical layers (McMurtrie and Wolf, 1983). The canopys radiation extinction coefcient (k), the most critical element of Beers law, dened as the average projection of leaves onto a horizontal surface, is a function of the area and form of the leaf, the leaf inclination, the zenith angle of the sun, and the leafs azimuth. In the case of the subordinate species, the use of the same k for both sole crop and intercrop stands in model development is inappropriate, as the canopy architectures of the understoreys in each case would be different (Graf et al., 1990; Wallace et al., 1991; Keating and Carberry, 1993). Nonetheless, some models have been designed that do not use the k of the partner species involved (Rimmington, 1984, 1985). Moreover, partitioning of radiation based on the vertical difference leads to a large underestimation (Wallace et al., 1991; Wallace, 1997; Tsubo and Walker, 2002), as the interception of radiation is independent of the stand height. Therefore, the past models have inherent limitations and thus warrant either further development or simplication. An improved model should be simple with few input variables for easy, rapid, and accurate estimations. The extinction coefcient computed from Beers law may provide accurate estimates of radiation interception by a sole crop or by the dominant crop of a mixture, which receive uniform direct solar rays, but it would probably not provide accurate measures of k or thereby prediction of light interception for a subordinate canopy, which experiences mostly diffuse radiation (Montieth and Unsworth, 1990; Campbell and Norman, 1998). Therefore, the k of a subordinate stand depends not only on its own canopy architecture and its optical properties but may also depend on some of the architectural and radiation characteristics of the dominant canopy. A new approach that takes this concept into account is needed. Based on the above facts, the present eld study was conducted to analyse radiation interception for use in exploring the yield advantage of a maize/peanut intercrop canopy, and to propose a simple model for the canopies of the separate layers in order to isolate the proportion of radiation intercepted by each partner on a daily scale. 2. Analysis of measurements Assuming that the canopy foliages are horizontally homogeneous, the amount of radiation intercepted (I)

76

M.A. Awal et al. / Agricultural and Forest Meteorology 139 (2006) 7483

by a crop stand generally obeys Beers law as I I 0 I g I 0 1 ekL ; (1)

canopy (I_) is simply I _ I I 0 1 ek ^ sole L ^ : (6)

where I0 and Ig are the radiant ux densities above and below (i.e., ground surface) the canopy, respectively, L the leaf area index (LAI) of the intercepted canopy, and k is the light extinction coefcient, representing the area of the shadow cast on a horizontal surface by the canopy divided by the plain area of the leaves in the canopy, which can be deduced from Eq. (1) as   Ig k ln =L; (2) I0 where the numerator following the natural logarithm indicates the fraction of the radiation transmitted ( f TR) by the canopy as f TR I g =I 0 : (3)

The amount of radiation transmitted through the dominant canopy (T^) is T ^ I 0 ek ^ sole L ^ : (7)

T^ is the incident radiation on the subordinate canopy. Therefore, I _ I 0 ek ^ sole L ^ 1 ek _ I _ Combining Eqs. (6) and (8) gives, I I 0 1 ek ^ sole L ^ I 0 ek ^ sole L ^ 1 ek _ L _ : (9) Therefore, k_ can be deduced as   Ig =L _ : k _ k ^ sole L ^ ln I0 3. Materials and methods 3.1. Field experiment A eld experiment was conducted at the farm of the Faculty of Agriculture, Niigata University, Japan (378580 N, 1398320 E) during the summer cropping season from May to October 2002. The basal fertilizers of N, P2O5, and K2O (4:8:15) were applied during the nal land preparation at a rate equivalent to 40 kg N ha1. Seeds of maize (cv. Cambel 78) and peanut (cv. Nakateyutaka) were planted by hand on 7 May in rows oriented east to west (EW). Three healthy seeds were planted per hole, and the seedlings were thinned to one vigorous plantlet soon after emergence. The crops were irrigated using the sprinkler method. Pest control and other cultural practices were performed to optimise growth and development with average maximum height about 160 cm for maize and 45 cm for peanut stands. 3.2. Experimental design The experiment comprised three treatments: sole crops of maize and peanut, and the maize/peanut intercrop, with three replicates. The row spacings were 50 and 100 cm for sole crops of peanut and maize, respectively. In the intercrop plot, the maize interrow distance was the same as for sole maize, and the peanut rows were grown between the maize rows, i.e., 50 cm from the maize rows. Therefore, both partners had equal (8)

(10)

In an intercropping system, the rows of dominant and subordinate species form the unit cell or intercrop prole. The row number of each partner in a unit cell depends mainly on the crop species, the growers demands, and the climatic conditions of the site. One fundamental condition is that all of the spaces in an intercrop cell be of equal weight during the radiation measurement, irrespective of the stand density (both the row number and the inter/intra-row spacing) of each partner. Let the intercropped canopy strata be separate for both partners, in other words, two vertically separated layers with the leaves of the intercepting canopy assumed to be horizontally homogeneous within each layer. It is assumed that the subordinate companion cannot affect the canopy geometry of the overstorey. Henceforth, the extinction coefcient of the intercropped dominant canopy (k^) is the same as that of the same species grown alone (k^sole). Therefore, the amount of radiation intercepted by the intercropped dominant canopy (I^) is I ^ I 0 1 ek ^ sole L ^ : (4)

If the k^sole is unknown, the dominant species should be grown alone to measure its extinction coefcient. The total radiation intercepted by the intercrop mixture (I^) is I I ^ I _ I0 Ig (5)

where subscripts ^, _, and ^ refer to the dominant and companion species and the combined intercropped stands. Therefore, the interception by the subordinate

M.A. Awal et al. / Agricultural and Forest Meteorology 139 (2006) 7483

77

population pressure. The intra-row spacing for all crops was 20 cm. Therefore, the stand density was 5 plants m2 for sole maize and 10 for sole peanut and the intercrop. 3.3. Data collection The radiation transmission through the canopy was recorded with a 1 m-long line quantum sensor (Li-Cor, Lincoln, NE, USA). A set of readings was taken from each plot at solar noon, including the incident photosynthetically active radiation (PAR) ux density below and above the canopy. Each value was the average of six successive readings. All sets of measurements were taken under clear-sky conditions during a period of constant irradiance. Data were recorded at intervals of about 817 days, depending on the sky conditions, starting from 5 June (30 days after planting) until the crops were mature. The line sensor was set perpendicularly to the crop row at the soil surface. The fraction of the PAR transmitted (f TPAR) was calculated using Eq. (3), and the fraction of the PAR intercepted (f IPAR) as 1 f TPAR. For the intercrop, an additional set of PAR measurements was taken at the same location after the removal of the peanut plants. The measurements were taken as rapidly as possible. The latter measurement was only for intercrop maize, and therefore, the interception by the intercrop peanut canopy could be obtained from the residual of these of two. The spatial transmission of the PAR in the horizontal plane of the row prole was measured on the same days by positioning the line quantum sensor parallel to the crop rows. The data were recorded at 5 cm intervals, dening the position at the south side of the prole of row stands running EW as 0 cm. P The total PAR intercepted I PAR between two consecutive measurements was calculated as (Awal and Ikeda, 2003) X I PAR Vb   f IPAR;1 f IPAR;2 =2; (11)

sampling. The leaf area was calculated from the relationship between the leaf dry weight (DW) and the specic leaf area, which were derived separately for each plot and date. The LAI was calculated as the leaf area/ground area (m2 m2). The harvested material was dried to a constant weight in a fan-forced oven at 80 8C. The total shoot DW was then recorded as the sum of the weights of the stem, leaves, and cobs (for maize) or the pods (for peanut). 3.4. Computation of k and e The temporal k value was determined using Eq. (2). For the season-long mean value, the k was computed as the slope of the linear regression forced through the origin from the numerator of Eq. (2) plotted against the corresponding denominator. Similarly, Eq. (10) was used to determine the temporal and seasonal mean k of intercrop peanut (i.e., k_). The data from Figs. 1 and 2 were used as input variables to compute the extinction coefcients for all treatments. The e of the seasonal changes (i.e., temporal) was calculated as the difference in the shoot dry weight between two consecutive measurements divided by the corresponding amount of PAR intercepted. The seasonal mean e was computed as the slope of the linear regression of the cumulative biomass plotted against the corresponding amount of PAR intercepted.

where V is the summation of the total daily incident shortwave solar radiation (Rs; 0.33.0 mm) between the rst (1) and second (2) investigation periods, and b (=0.45) is the conversion factor of Rs, to yield the PAR at 0.40.7 mm (Monteith, 1977). The daily incident Rs ux was recorded at the onsite Agriculture Meteorological Station (Shindhori, Niigata). Soon after each radiation measurement, all plants within an area of 1 m2 were harvested for destructive

Fig. 1. Leaf area index (LAI) as a function of time. Vertical bars indicate the standard deviation of mean values from the three replications.

78

M.A. Awal et al. / Agricultural and Forest Meteorology 139 (2006) 7483

4.3. Spatial transmission of PAR The spatial transmission of PAR across the row with different crop canopies is shown in Fig. 3. In sole maize (Fig. 3a), until 5 June, all the radiation reached the ground, i.e., f TPAR equalled unity for the part 2080 cm between the rows. As canopy development progressed, the row structure was clearly evidenced through a decrease in the f TPAR. The f TPAR decreased until 25 July, followed by a gradual increase toward maturity. In sole peanut (Fig. 3b), the f TPAR was initially at maximum with almost no row structure. In subsequent stages, the row structure appeared to result in a nearly normal frequency distribution about the position of the peanut plants. In intercrop canopies (Fig. 3c), the variation across the row was similar to that of sole maize. Along the peanut row (50 cm), the only difference from sole maize on 5 June was slightly lower transmission, i.e., a slight decrease in transmission between 45 and 55 cm differentiated it from sole maize, which is the main effect on companion stands between maize rows in intercrop associations. Except for just above the peanut row, both the increase and decrease in seasonal transmission followed the pattern of maize alone. The f TPAR was higher at the north end of the inter-row prole than at the south end in all canopies (Figs. 3a and c, and 4) except for that of sole peanut (Fig. 3b). The spatial f TPAR through the intercrop peanut canopy following the maize harvest also showed a distinct row structure (Fig. 4). With further growth, the f TPAR gradually decreased toward the maturity of the plants on 3 October. 4.4. Canopy light extinction coefcients (k) The evolution of k values over a season is shown in Fig. 5. Initially, the canopies of all of the treatments had low extinction coefcients, followed by higher values at subsequent stages. At all growth stages, the canopies of the sole crop and the intercrop of maize had similar and stable k values that were lower than those of both the sole crop and the intercrop of peanut, except for the rst two measurements. The k value of the sole peanut crop sharply increased with time to form a peak on 25 July, and declined rapidly thereafter. The k value of intercrop peanut was slightly higher than that of sole peanut until 6 July, but then dropped suddenly and continued to decrease steadily for the remainder of the intercropping period. The intercrop peanut stands had lower k values than their sole crop canopy at the time of the most

Fig. 2. Fraction of PAR intercepted ( fIPAR) as a function of time. Vertical bars are as in Fig. 1.

4. Results 4.1. Meteorological parameters A summary of the meteorological data across the study period is presented in Table 1. Although the air temperature gradually increased throughout the growing season, the relative humidity (RH) was nearly unchanged in all months except July, when the RH was higher due to heavy precipitation (386 mm) that suppressed the hours of light and the incident solar radiation. 4.2. Temporal leaf area development and PAR interception In sole maize, the LAI rapidly increased to a maximum on 16 July, at about 2.5 m2 m2 (Fig. 1). A LAI nearly as high (at least 2.0 m2 m2) was maintained from 6 July to 7 August. The LAI of intercrop maize was slightly lower than that of sole maize, but followed the same pattern. The LAI of sole peanut increased gradually until 25 July, followed by a sharp increase to a peak of about 3.5 m2 m2 on 26 August. Subsequently, the LAI decreased as the plants approached maturity on 3 October. The LAI of intercrop peanut stands increased slowly until the harvest of the partner maize on 26 August, followed by a rapid increase. Over the whole cultivation period, the fraction of PAR intercepted followed almost the same pattern as the LAI in all canopy types (Fig. 2).

M.A. Awal et al. / Agricultural and Forest Meteorology 139 (2006) 7483 Table 1 Meteorological parameters at the onsite agricultural station during the growing period from May to October 2002 Month (2002) Air temperature (8C) Mean maximum 13 18 22 23 19 15 18 (4) Mean minimum 3 7 12 13 8 4 8 (4) Relative humidity (%) Mean maximum 69 68 77 72 71 73 72 (3) Mean minimum 46 47 57 50 46 50 49 (4) Total precipitation (mm) 93 51 386 120 67 147 144 (124) Average light duration (h day1) 9 10 8 9 7 6 8 (2)

79

Total shortwave solar radiation (MJ m2) 477 504 398 468 331 179 393 (122)

May June July August September October Mean ()

Values in parentheses represent the standard deviation () of the values for all months.

effective intercropping period (between 6 July and 18 August) when the dominant maize offered more shade and had the maximum LAI. Notably, the k value for intercrop peanut showed a larger standard deviation

than those of other canopies. The seasonal mean k values are presented in Table 2. The canopies of sole and intercrop maize exhibited similar k values. In contrast, the k value of intercropped peanut was about 17% lower than that of sole peanut, with a larger variability than for other canopies. 4.5. Seasonal changes in the e and mean e In sole maize over the whole season, the e never dropped below 2 g(DW) MJ1, except during the week prior to the nal harvest, when the e became negative (Table 3). The e was maximal from 19 June to 6 July, at 5.81 g(DW) MJ1. The seasonal mean e also had a higher value, at 3.27 g(DW) MJ1. Both the seasonal changes and the mean e of intercropped maize were very similar to those of the stands with sole crop. Initially, the e of sole peanut was somewhat lower, but soon reached a

Fig. 3. Fraction of PAR transmitted ( fTPAR) through the different canopies. (b) Two horizontal proles of three consecutive rows of sole peanut. Vertical bars are as in Fig. 1.

Fig. 4. Fraction of PAR transmitted ( fTPAR) through the intercrop peanut canopy following the maize harvest. Vertical bars are as in Fig. 1.

80

M.A. Awal et al. / Agricultural and Forest Meteorology 139 (2006) 7483 Table 2 Seasonal mean light extinction coefcient (k) of canopies under different cropping systems Crop canopy Maize Sole crop Intercrop Peanut Sole crop Intercrop Light extinction coefcient (k) 0.441 0.008 b 0.434 0.015 b 0.599 0.030 a 0.496 0.061 b

Values represent the standard deviation of three replicates. Values followed by different letters within a column are signicantly different (P < 0.01).

4.6. Yield potentiality The grain yield of intercropped maize was slightly less than that of stands with sole maize (Table 4). Intercrop peanut plants yielded about 46% of the pods produced by peanut grown alone. The grain and pod yields of the combined intercrop were 44 and 31% greater than those of maize and peanut alone, respectively. Although there was no variation in the HI between sole and intercrop maize, the HI was about 13% lower in intercropped peanut than in stands grown alone. 5. Discussion The consecutive rows of maize foliage occupied most of the interrow space at the period of maximum vegetative development, leaving a narrow opening over the peanut row (Fig. 3c). As a result, the peanut

Fig. 5. Light extinction coefcient (k) as a function of time. Vertical bars are as in Fig. 1.

maximum on 6 July, at 3.42 g(DW) MJ1. Thereafter, the e gradually declined until 26 August, followed by an increase during the subsequent pod-lling stage. The seasonal changes in the e of intercropped peanut followed a pattern very similar to that of the sole crop stands, but with higher magnitudes. The season-long e of intercropped peanut was about 79% higher than that of sole crop. The e for the combined intercrop changed continuously over the growth stages. However, the seasonal mean e of the combined intercrop was 7% lower than that of maize, but more than two-fold higher than that of sole peanut.
Table 3 Radiation-use efciencies, e (g(DW) MJ1), for different cropping systems Day/month (2002) Maize Sole crop Seasonal changes (temporal) 519 June 19 June to 6 July 616 July 1625 July 25 July to 7 August 718 August 1826 August 26 August to 10 September 1019 September 19 September to 3 October Seasonal mean (regressed data) 5 June to 26 August 5 June to 3 October 2.15 5.81 4.56 3.85 2.65 2.27 0.32 3.27 Intercrop 2.16 6.16 4.32 3.25 3.05 1.85 0.27 3.26

Peanut Sole crop 1.54 3.42 3.14 2.15 1.22 1.17 0.60 0.69 1.68 1.42 1.44 1.19 Intercrop 2.63 5.15 3.47 3.68 2.15 0.88 0.63 1.41 1.64 1.55 1.59 2.13

Intercrop combined

2.25 5.99 4.19 3.32 2.90 1.62 0.18 3.03

M.A. Awal et al. / Agricultural and Forest Meteorology 139 (2006) 7483 Table 4 Grain and pod yield and harvest index at physiological maturity for different cropping systems Cropping system Maize Sole crop Intercrop Peanut Sole crop Intercrop Combined intercrop Grain or pod yield (g m2) 272 12 b 255 21 b 299 28 b 136 07 c 392 14 a Harvest index (%) 32 0 c 32 0 c 55 4 a 48 2 b

81

Values represent the standard deviation of three replicates. Values followed by different letters within a column are signicantly different (P < 0.01).

plants formed clustered and erect canopies (Smith, 1982). Simultaneously, as canopy development progressed, the shading by the taller maize increased, which increased the ratio of diffuse to direct radiation inside the intercrop canopy. This higher ratio beneath the maize canopy probably contributed to the lower k of intercropped peanut. The erect nature of the subordinate peanut foliage is another factor in the lower k value of these plants. In all plant types, an initial higher evolution of the k indicates a rapid projection of the canopy elements onto the horizontal surface to trap the maximum amount of radiation after seedling establishment. The higher k of both sole and intercrop peanut compared to that of maize stands can be explained by the fact that the peanut canopy expands more slowly than that of maize (Black and Ong, 2000). The sudden drop in the k of intercrop peanut after 6 July may have been caused by the rapid ush in the ratio of diffuse to direct radiation as the LAI of the dominant maize reached a maximum. Similarly, the k of this canopy slowly decreased with time due to the increasing shade from the dominant maize canopy. In contrast, the rapidly declining k of the sole peanut canopy after 25 July indicates that the PAR interception reached its maximum potential in spite of the LAI increasing subsequently. The larger variability in the k of intercrop peanut represents its higher sensitivity to the magnitudes of shade offered by the taller maize plants. The similar k values for sole and intercrop maize indicate that their canopy geometries were similar due to experiencing almost the same radiation environment. The very similar k values of sole and intercrop maize support the assumption (Section 2) that the canopy geometry of dominant maize plants is largely unaffected by the subordinate partner. Hence, the proposed model is not only applicable for the maize/peanut

intercrop, but also for other canopy mixtures in which one partner (dominator) is either not affected or only marginally affected by the subordinate companion. However, this simple model needs to be investigated further using other species and canopy types. The slowly developing subordinate canopy does not intermingle with the dominant canopy due to the senescence of leaves of the lower portion; thus, separate strata are maintained for the whole intercropping period. Some examples of mixtures that form separate strata in intercrop associations are potato, Phaseolus bean, Stylosanthes, lentil, Lathyrus, and dwarf tomato with maize or sugarcane, and peanut with pearl millet (Keating and Carberry, 1993, and references therein). Additionally, this principle would be of much interest for agroforestry systems in which annual crops form separate strata beneath perennial trees. However, it may not be possible to partition the radiation intercepted by plants in mixtures in which the canopy foliage of the partners intermingle. The higher ratio of diffuse to direct radiation surrounding the intercrop peanut would help to improve its e value (Sinclair et al., 1992; Healey et al., 1998). The shading effect offered by the taller maize might be partially offset by its higher photosynthetic rates per unit PAR at low intensity (Warren Wilson, 1969). The higher e of intercropped peanut can also be explained by its low k value, which provides a valuable signature indicating that this canopy intercepts less PAR per unit DW produced, enhancing the e. The e of intercrop peanut was also reported to be boosted when grown with millet (Marshall and Willey, 1983) and sorghum (Harris et al., 1987). Our results corroborate these reports. The higher light transmission at the north side of the interrow prole can be explained by the relationship between the zenith of solar rays and the direction of the row. Because the experimental site was at about 388N, the southern side of the across row proles would be shaded by the maize plants. Some open spaces at the centre between EW-oriented rows allowed such inclined rays, resulting in more transmission to the north side. The dense canopy of a sole peanut stand could eliminate such spatial transmission.The seasonal leaf area development, the radiation intercepted by the canopy, and the spatial transmission indicate explicitly that there was ample free space between the interrows of maize until 6 July, the rst 2 months of growth. In sole maize, the radiation that reaches this space might not be utilised within that period. Subsequently, the space was occupied by the developing canopy from 6 July to 7 August, followed by a reopening of the area due to leaf senescence. Therefore, the subordinate peanut would

82

M.A. Awal et al. / Agricultural and Forest Meteorology 139 (2006) 7483

experience four distinct growth phases (Figs. 1 and 2): rst, an initial normal rate of development following seedling emergence; second, an interim lag phase of growth retardation from 6 July to 7 August while being suppressed by the dominant maize; third, commencement of later growth following the lag period up to the maize harvest; fourth, strong growth while growing alone following the maize harvest. It may be difcult to dene the transition between the third and fourth phases. As the dominant partner, the maize plants grow independently of the companion peanut plants. The subordinate peanut plants grow in a compatible manner rather than competing, exploiting the remainder of the resources between the maize rows. As a result, the yield of intercropped maize is not signicantly inhibited. The shade effect on the peanut canopy during the interim lag period due to the taller maize is then compensated for by the full radiation following the maize harvest, which results in substantial production of peanut pods. These pods could represent an additional yield for this cropping system obtained because of the efcient utilisation of natural resources such as light. Other studies have also found such yield advantages with subordinate peanut intercropped with taller cereals such as millet (Reddy and Willey, 1981) and sorghum (Harris et al., 1987). The lack of sufcient radiation during the middle stage of growth (the lag phase) could not be fully overcome in terms of the biomass partitioning toward the pods, resulting in a reduced HI for intercrop peanut. This nding contradicts the results of the above citations, but concurs with those on the study of a maize/bean intercropping, in which the HI was reduced in the subordinate bean (Tsubo et al., 2001). The canopy architecture of the upperstorey and the relative contribution of each partner in the intercrop prole may be causes of such varying ndings. 6. Conclusions For intercropping, although the light extinction coefcient of the dominant maize is unaffected but that of the subordinate peanut is signicantly reduced, a simple radiation partitioning model is proposed which can be used for wide range of stand densities. The model may also be applicable in agroforestry systems. Peanut plants can tolerate the shade produced by maize plants when grown as an intercrop. Since the peanut season is longer than that of maize, in localities where a fallow period normally follows the maize harvest, intercropped peanut would produce some pods, a great advantage for growers. However, growers must choose appropriate cultivars to best synchronise the

crops. The higher yields obtained in intercropping systems are probably due to greater radiation-use efciencies. Acknowledgements The senior author was supported by the Jin-nai International Student Scholarship Program. We would like to thank Professor Takeshi Ohta of Nagoya University for his cooperation. We are grateful to Dr. J.B. Stewart and anonymous reviewers for fruitful comments and suggestions to improve the nal version of the manuscript. We thank the editors at Textcheck, a scientic and technical editing service (http://www. textcheck.com/), for helping us revise this manuscript. References
Awal, M.A., Ikeda, T., 2003. Effect of elevated soil temperature on radiation-use efciency in peanut stands. Agric. For. Meteorol. 118, 6374. Black, C., Ong, C., 2000. Utilization of light and water in tropical agriculture. Agric. For. Meteorol. 104, 2547. Campbell, G.S., Norman, J.M., 1998. An Introduction to Environmental Biophysics, 2nd ed. Springer-Verlag Inc., New York. Connolly, J., Goma, H.C., Rahim, K., 2001. The information content of indicators in intercropping research. Agric. Ecosys. Environ. 87, 191207. Corlett, J.E., Black, C.R., Ong, C.K., Monteith, J.L., 1992. Above- and below-ground interactions in a leucaena/millet alley cropping system. II. Light interception and dry matter production. Agric. For. Meteorol. 60, 7391. Graf, B., Gutierrez, A.P., Rakotobe, O., Zahner, P., Delucchi, V., 1990. A simulation model for the dynamics of rice growth and development. Part II. The competition with weeds for nitrogen and light. Agric. Syst. 32, 367392. Harris, D., Natarajan, M., Willey, R.W., 1987. Physiological basis for yield advantage in a sorghum/groundnut intercrop exposed to drought. 1. Dry-matter production, yield, and light interception. Field Crops Res. 17, 259272. Healey, K.D., Rickert, K.G., Hammer, G.L., Bange, M.P., 1998. Radiation use efciency increases when the diffuse component of incident radiation is enhanced under shade. Aust. J. Agric. Res. 49, 665672. Keating, B.A., Carberry, P.S., 1993. Resource capture and use in intercropping: solar radiation. Field Crops Res. 34, 273301. Marshall, B., Willey, R.W., 1983. Radiation interception and growth in an intercrop of pearl millet/groundnut. Field Crops Res. 7, 141160. McMurtrie, R., Wolf, L., 1983. A model of competition between trees and grass for radiation, water and nutrients. Ann. Bot. 52, 449458. Monteith, J.L., 1977. Climate and efciency of crop production in Britain. Philos. Trans. R. Soc. Lond. B 281, 277294. Montieth, J.L., Unsworth, M.H., 1990. Principles of Environmental Physics, 2nd ed. Edward Arnold, UK. Ofori, F., Stern, W.R., 1987. Cereal-legume intercropping systems. Adv. Agron. 41, 4190.

M.A. Awal et al. / Agricultural and Forest Meteorology 139 (2006) 7483 Reddy, M.S., Willey, R.W., 1981. Growth and resource use studies in an intercrop of pearl millet/groundnut. Field Crops Res. 4, 1324. Rimmington, G.M., 1984. A model of the effect of interspecies competition for light on dry-matter production. Aust. J. Plant Physiol. 11, 277286. Rimmington, G.M., 1985. A test of a model for light interception by mixtures. Aust. J. Plant Physiol. 12, 681683. Sinclair, T.R., Shiraiwa, T., Hammer, G.L., 1992. Variation in crop radiation-use efciency with increased diffuse radiation. Crop Sci. 32, 12811284. Sinoquet, H., Bonhomme, R., 1992. Modeling radiative transfer in mixed and row intercropping systems. Agric. For. Meteorol. 62, 219240. Sinoquet, H., Rakocevic, M., Varlet-Grancher, C., 2000. Comparison of models for daily light partitioning in multispecies canopies. Agric. For. Meteorol. 101, 251263.

83

Smith, H., 1982. Light quality, photoperception, and plant strategy. Ann. Rev. Plant Physiol. 33, 481518. Tsubo, M., Walker, S., Mukhala, E., 2001. Comparisons of radiation use efciency of mono-/inter-cropping systems with different row orientations. Field Crops Res. 71, 1729. Tsubo, M., Walker, S., 2002. A model of radiation interception and use by a maizebean intercrop canopy. Agric. For. Meteorol. 110, 203215. Wallace, J.S., 1997. Evaporation and radiation interception by neighbouring plants. Q.J.R. Meteorol. Soc. 123, 18851905. Wallace, J.S., Batchelor, C.H., Dabeesing, D.N., Teeluck, M., Soopramanien, G.C., 1991. A comparison of the light interception and water use of plant and rst ratoon sugar cane intercropped with maize. Agric. For. Meteorol. 57, 85105. Warren Wilson, J., 1969. Maximum yield potential. In: Proceedings of the VIIth Colloquium International Potash Inst, Bern, Switzerland, pp. 3456.