| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Central Agric. Res. Center, Montana State Univ., HC90 Box 20, Moccasin, MT 59462
b Western Agric. Res. Center, Montana State Univ., 580 Quast Lane, Corvallis, MT 59828
* Corresponding author (cchen{at}montana.edu)
Received for publication January 7, 2004.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSNOTESRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: ADF, acid detergent fiber • ANOVA, analysis of variance • CARC, Central Agricultural Research Center • LER, land equivalent ratio • MS, mean square • NDF, neutral detergent fiber • RCB, randomized complete block • WARC, Western Agricultural Research Center
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSNOTESRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Total forage production of an intercropping system is dependent on the enhancement or suppression of each species. Interspecies competition for growth resources, such as water and light, can result in suppression of growth and biomass accumulation in the less competitive species. Reports in the literature show contradictory intercropping effects on total forage yield (Robinson, 1960; Walton, 1975; Carr et al., 1998; Chapko et al., 1991). Robinson (1960) reported that pea improved oat (Avena sativa L.) forage yield. In a 2-yr pea–barley and pea–oat intercropping study, Carr et al. (1998) found that total forage yield was unaffected by intercropping when the cereal crop was sown at a rate equal to or greater than the sole crop seeding rate. However, less forage was produced when the cereal component was sown at half the sole crop seeding rate. They also found that the intercropping forage yield was unaffected by the pea seeding rate. In other studies, forage and grain yield of legumes were suppressed by cereal components (Ofori and Stern, 1987; Hauggaard-Nielsen and Jensen, 2001; Hauggaard-Nielsen et al., 2001). Seeding rates for component crops in cereal–pea mixtures are commonly less than when either the cereal crop or pea is sown alone (Carter and Larson, 1964; Droushiotis, 1989).
The efficiency of an intercropping system can be evaluated by the land equivalent ratio (LER), defined as the total area required under sole cropping to produce the equivalent yields obtained under intercropping (De Wit and Van Den Bergh, 1965; Willey, 1979; Mohta and De, 1980). It is expressed as:
 | [1] |
In an intercrop system, row configurations (arrangements) alter the amount of light transmission to lower layers of the crops and affect the competition of species for light, water, and nutrients. There are four types of intercropping row configurations: (i) mixed intercropping, which grows component crops simultaneously in complete mixtures; (ii) row intercropping, which grows component crops simultaneously in different rows; (iii) strip intercropping, which grows component crops simultaneously in different strips; and (iv) relay intercropping, which grows component crops in relay so that growth cycles overlap (Sullivan, 2001).
In a maize (Zea mays L.)–pigeon pea [Cajanus cajan (L.) Millsp.] system, Dalal (1974) found that maize yield was unaffected in an alternate row configuration (row intercropping) but was reduced when both species were planted in the same row (mixed intercropping). In maize–soybean [Glycine max (L.) Merr.] and sorghum [Sorghum bicolor (L.) Moench]–soybean intercropping systems, Mohta and De (1980) reported that yields of maize and sorghum were unaffected in either single or double alternate row configurations. However, there was a 31% soybean yield increase in a maize–soybean system and a 26% increase in the sorghum–soybean system when components were arranged in double alternate rows compared with single alternate rows.
Intercropping cereal and legume species may improve the efficiency of N use. The cereal component usually has faster growing or more extensive root systems than the legume component and is more competitive for soil inorganic N (Anil et al., 1998; Carr et al., 1998; Carruthers et al., 2000). This forces the legume component to fix N from the atmosphere (Jensen, 1996; Hauggaard-Nielsen et al., 2001). Pea dry matter was reduced by fertilizer N in a barley–pea mixture, which was attributed to the vigorous growth and strong competitiveness of barley stimulated by the fertilizer (Anderson et al., 1983). Jensen (1996) also found that fertilizer N did not influence the total intercrop yield, but decreased the proportion of pea contribution. In a maize–soybean intercrop system, Dalal (1977) found reduced seed yield of soybean with increased N addition. The LER was reduced from 1.15 without N application to 1.09 with 100 kg N ha–1 applied. In a similar study, Chui and Shibles (1984) found that fertilizer N increased maize grain yield but decreased soybean seed yield. A study using 15N isotope indicated that no significant direct belowground N transfer from legume to grass was observed during the lifetime of the legume; however, after cutting the shoot of the legume at ground level, the grass assimilated significant amounts of N derived from decaying legume roots (Trannin et al., 2000).
Types of row configurations and N application strategies to obtain a good quality and high yield forage in barley–pea intercropping systems have not been well documented, particularly in Montana. The objectives of this study were to determine the effects of (i) row configuration and (ii) fertilizer N on forage yield, protein content, and LER of barley–pea intercropping systems.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSNOTESRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Intercropping Systems
A commercial hay barley cultivar Haybet and an experimental selection of Austrian winter pea MocSelect92 were intercropped in 2000, 2001, and 2002 at WARC and CARC with three different intercrop row configurations and three fertilizer N levels. The first intercrop row configuration consisted of four rows of barley planted adjacent to four rows of pea (4 x 4). The second row configuration consisted of two rows of barley planted adjacent to two rows of pea (2 x 2) and the third row configuration consisted of barley and pea mixed within the same row (mixed). At WARC, the trial was planted into tilled buckwheat (Fagopyrum esculentum Moench) stubble in all 3 yr. A 1.2-m wide four-row planter with a separate seed cone for each row was used to seed the plots. The planter was equipped with hoe openers at a 0.30-m row spacing. Plot dimensions were 1.2 by 4.6 m (one drill pass) for the mixed and 2 x 2 row configurations, and 2.4 by 4.6 m (two drill passes) for the 4 x 4 row configuration. Two rows of barley or pea borders were planted next to each side of a treatment plot (i.e., barley border was planted next to pea and pea border was planted next to barley) in the 2 x 2 and 4 x 4 row configurations. The plots were planted in late April and were irrigated during the period between May and June in 2000 and 2001. Due to the dry spring conditions, irrigation was applied in April and May in 2002. Approximately 178, 127, and 76 mm of water were applied in 2000, 2001, and 2002, respectively. Precipitation during the growing season was 45, 95, and 108 mm in 2000, 2001, and 2002, respectively.
The trial was planted into tilled winter wheat stubble in 2000 and 2001 and no-tilled in 2002 at CARC. A 1.8-m wide 6-row planter with a separate seed cone for each row was used to seed the plots. The planter was equipped with disc openers with a 0.30-m row spacing. Plot dimensions were 1.2 by 7.6 m for the mixed and 2 x 2 row configurations, and 2.4 by 7.6 m for the 4 x 4 row configuration. The trials were planted in late April of each year. Irrigation was not applied at this site. Precipitation during the growing season was 156, 141, and 158 mm in 2000, 2001, and 2002, respectively. At both locations for all 3 yr, peas were inoculated with Nitrastik-C1 (Liphatech, Milwaukee, WI) granular inoculant (Rhizobium Leguminosarum bv viceae) before seeding.
The seeding rate for the pea component was 75 seeds m–2 in the 2 x 2 and 4 x 4 row configurations and 43 seeds m–2 in the mixed. The seeding rate for the barley component was 161 seeds m–2 in the 2 x 2 and 4 x 4 row configurations and 65 seeds m–2 in the mixed. The seeding rates in the 2 x 2 and 4 x 4 configurations were calculated based on the area with pea or barley pure stands, but the seeding rate for the mixed configuration was calculated based on the area with pea–barley mixed stands (intercrops). The seeding rates adapted in this study were based on previous field studies on pea and barley sole- and inter-cropping at Montana State University (Dennis Cash, personal communication, 2000). The barley and pea seeding rates used in this study were similar to the recommended seeding rates for sole crops (185 and 80 seeds m–2, respectively) that Carr et al. (1998) used in North Dakota.
Fertilizer Nitrogen
Soil samples from each experiment site were taken before seeding to test the background nutrition level. Because of the continuous cropping every year, the residual N was <30 kg ha–1 in 0 to 60 cm soil in all sites and years. The experimental sites were responsive to N application. There were three N levels for each row configuration: N0, N1, and N2 representing 0, 67, and 134 kg N ha–1, respectively. Fertilizer N was spread by hand in mid-May at the two to three leaves stage of the barley using ammonium nitrate. At WARC, the N was applied only to the barley component in the 2 x 2 and 4 x 4 alternate row configurations, but was applied to both pea and barley components in the mixed configuration. At CARC, the fertilizer N was applied to both barley and pea components in all row configurations in 2000 and 2001, but the fertilizer application strategy was the same as at WARC in 2002. Fertilizer P and K were sufficiently applied before the experiment was established according to the recommendation for field production in the region. The experiment was arranged as a two-factorial randomized complete block (RCB) design with four replications. Row configurations and N treatments were completely randomized within each block.
Barley and pea forage were hand harvested at late anthesis to early milk stage (Zadoks Growth Stages 69 to 74; Zadoks et al., 1974) of the barley component, and at flowering to pod-setting at lower nodes of the pea component (Pea Growth Stage 203 to 205; Knott, 1987) in late June to early July. Two rows of barley and pea by plot length were harvested from each plot in the 2 x 2 row configuration. In the 4 x 4 row configuration, four rows by plot length were harvested and the two middle rows were separated from the two outer rows, or edge rows (rows next to the other species), to evaluate species competition. In the mixed configuration, two center rows by plot length were cut and species components were separated in the field. After fresh weights of each component crop were recorded, forage samples were placed in a dryer at 70°C until completely dry (at least 72 h) and the dry weights recorded. A subsample was taken and ground for protein analysis. Additional forage samples from WARC were taken to analyze for NO–3, NDF, and ADF contents in 2001, and only for NO–3 in 2002. These analyses were conducted at WARC. Total N in plant tissue was determined by the micro-kjeldahl method and converted to protein content (Jones and Case, 1990). The NO–3 was extracted with aluminum sulfate and measured by an ion-specific electrode (Griffin, 1995). The NDF and ADF were measured using an ANKOM Fiber Analyzer System (ANKOM Technology, 2003).
Yield Calculation and Statistics
Dry weights of barley and pea samples were converted to the biomass yield for each component crop. Yield of barley (Yb) and pea (Yp) from each plot were added together to make the total biomass yield (Yt). Middle rows in strip intercrops were used to represent sole crop yield in previous research (Lesoing and Francis, 1999; Ayisi et al., 1997), so barley and pea yield from the two middle rows in the 4 x 4 row configuration was used to estimate sole crop yields when computing the LER. The intercropping effect on middle rows of pea and barley was considered minimal since plants were harvested before soil water was depleted to greater depths by barley than pea, and because shading differences between crop species was unlikely since plant height was similar. The sole crop yield was calculated at each of the three N levels in this study. The LER for the barley and pea components—as well as for the pea and barley combined intercrops—were calculated using Eq. [1]. The LER was calculated for both biomass and protein bases.
Analysis of variance (ANOVA) for RCB design was performed to determine row configuration and fertilizer N effects on barley and pea components and combined biomass yield (Yb, Yp, and Yt), protein content (Pb, Pp, and Pt), and LER on a biomass yield (Lb-Y, Lp-Y, and Lt-Y) and protein basis (Lb-P, Lp-P, and Lt-P). Because of the differences in soil type and irrigation at the WARC and CARC sites, ANOVA was conducted separately for each site. A mixed model was used to analyze combined multi-year data at WARC and CARC. The year was treated as a random effect nested under replications (blocks), while row configuration and N were treated as fixed effects. The F test for year effect was done by MSyear/MSblock/year (MS is mean square), row configuration effect was tested by MSrow/MSrow x year, N effect was tested by MSN/MSN x year, and row x N interaction was tested by MSrow x N/MSrow x N x year. The interactions of row x year, nitrogen x year, and row x nitrogen x year were tested against the residual error (McIntosh, 1983; Carr et al., 2004). If the ANOVA for the multi-year combined data showed significant interaction effects between treatments and year, separate ANOVA was conducted for each individual year. Main treatment means were separated and tested by Fisher's protected LSD at P = 0.05 significance level. The statistical package SYSTAT 10.2 (SYSTAT Software, Richmond, CA) was used for all analyses.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSNOTESRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
At CARC, the biomass yield of the barley component was the greatest in the mixed configuration, followed by the 2 x 2 and 4 x 4 configurations, respectively. Similar to the WARC results, the biomass yield of pea was reduced by interspecies mixing during the 3 yr. The yield of pea in the 4 x 4 configuration was greater than in the 2 x 2, followed by the mixed. Total combined forage yield was greater in the mixed than in the 4 x 4 configuration.
Although barley biomass yields did not differ between the N1 and N2 levels, yields at these two levels were greater than at the N0 level (Table 1). The yield of the pea component was unaffected by fertilizer N, and the total combined biomass yield at N1 and N2 levels was greater than at the N0 level. Dryland forage production at the CARC site was less than the irrigated forage production at WARC. Barley at CARC produced 50% less biomass than at WARC.
Protein Content
The ANOVA for combined data across the years did not show a significant row configuration effect on Pb during the 3-yr study at WARC (Table 2), but the ANOVA for individual years indicated that the mixed row configuration increased barley protein content in 2000 and 2001 but not in 2002 (data not shown). The protein content of the pea component was less in the mixed than in the 2 x 2 and 4 x 4 configurations. The protein content of barley–pea combined forage did not differ among the row configurations. Barley protein content ranged from 94 to 106 g kg–1, and pea protein ranged from 167 to 195 g kg–1. The protein content of the combined forage ranged from 114 to 119 g kg–1 in the three intercropping row configurations.
|
Row configuration at CARC did not affect barley and pea protein content. Protein content of the combined for age was greater in the 4 x 4 than in the 2 x 2 and mixed configurations (Table 2), but the magnitude of the differences were small. Mean Pt was 137 g kg–1 for the mixed, 140 g kg–1 for the 2 x 2, and 147 g kg–1 for the 4 x 4 row configurations. Protein content of the barley component ranged from 117 to 123 g kg–1, and the protein content of the pea component ranged from 193 to 195 g kg–1.
Barley protein content increased with increased N levels (N2 > N1 > N0). Pea protein content at the N2 level was greater than at the N1 and N0 levels. The protein content of the combined forage also increased with increased N levels (N2 > N1 > N0).
Land Equivalent Ratio on Biomass Basis
During the 3-yr study at WARC, row configurations that increased the interspecies mixture also increased Lb-Y, but did not affect Lp-Y (Table 3, Fig. 1a). The greatest Lb-Y value was obtained in the mixed configuration, followed by the 2 x 2 and 4 x 4 configurations, respectively. The Lt-Y was greater in the mixed than in the 2 x 2 and 4 x 4 configurations. Intercropping increased the use efficiency of plant growth resources by 5 to 24% (Lt-Y value ranging from 1.05 to 1.24), with the greatest growth resource use efficiency at the mixed configuration, for which the Lt-Y value was 1.24. The barley component was the dominant species in the intercropping systems, especially in 2 x 2 and mixed configurations where the Lb-Y values were 0.64 and 0.82, respectively. Pea growth was suppressed in the intercropping systems, particularly in the mixed configuration, for which the Lp-Y value was 0.42.
|
|
Interspecies mixing increased Lb-Y but decreased Lp-Y during the 3 yr at the CARC site (Table 3, Fig. 2a). The Lb-Y was the greatest in the mixed configuration, followed by the 2 x 2 configuration and the 4 x 4 configuration. However, Lp-Y was greater in the 4 x 4 and 2 x 2 than in the mixed configuration. The Lt-Y value did not differ among the row configurations. Similar to the results at WARC, the mean Lt-Y values ranged from 1.05 to 1.09, indicating the use efficiency of growth resources was increased by 5 to 9% in the intercropping systems. Barley was the dominant species with a Lb-Y of 0.79 in the mixed configuration. Pea was the less competitive component having Lp-Y value of 0.29 in the mixed configuration.
|
Land Equivalent Ratio on Protein Basis
Effects of main treatments on the LER calculated on a protein basis in Table 4 were similar to those calculated on a biomass basis in Table 3. At WARC, the Lb-P was the greatest in the mixed followed by the 2 x 2 and 4 x 4 configurations, respectively. However, the Lp-P was lower in the mixed than in the 2 x 2 and 4 x 4 row configurations. The Lt-P was greater in the mixed than in the 4 x 4 configuration.
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSNOTESRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Main treatment effects on protein content were inconsistent among years and sites. Although Pb increased with interspecies mixing at the WARC site in 2000 and 2001, it was unchanged in 2002. It was unaffected at CARC during the 3 yr. Application of N consistently increased Pb in all years and locations. However, intercropping and N effects on protein content of the pea component were inconsistent at both locations. Increasing pea–barley mixture decreased the Pp at the WARC site but did not affect the Pp at the CARC site. These differences likely resulted from the difference in production practices (irrigation vs. dryland) and soil depth at the two locations. Irrigation was applied at critical times and there was a deeper soil profile at the WARC site than at the CARC site (
1.0 m at WARC and 0.5 m at CARC), suggesting that the environment at the WARC site was more favorable than at the CARC site. The barley component at the WARC site produced twice as much biomass as at the CARC site, but the difference of pea biomass at the two locations was small compared with the barley component (Table 1). This implies that the barley component at WARC took up much more soil inorganic N than at CARC, and the competition for N from barley in the mixed configuration also was much stronger at WARC than at CARC. This competition could have resulted in the lower protein content of the pea component in the mixed configuration. Other researchers also suggested that barley had more vigorous vegetative growth and stronger competition than pea for soil inorganic N (Anderson et al., 1983; Anil et al., 1998; Jensen, 1996). The protein content of combined forage Pt was higher at CARC (137–147 g kg–1) than at WARC (114–119 g kg–1) because the pea component at CARC made up a greater proportion of the forage (13–24% at WARC vs. 19–37% at CARC; Table 1).
Although the N was applied differently between the first 2 yr and the third year in the 2 x 2 and 4 x 4 row configuration treatments at CARC (i.e., fertilizer N was applied to pea component in the first two years but not in the third year), there was no strong evidence indicating these application methods affected the biomass yield and protein content of the pea component. The biomass of pea was unaffected by N, but protein content was increased every year.
Whether or not an intercropping system is superior to sole cropping depends on the enhancement and suppression of each component crop in the intercrops when compared to sole crops. In this study, intercrop biomass and protein production exceeded sole cropping (i.e., the LER on biomass and protein bases exceeded 1.0), indicating an advantage of intercropping, particularly in the mixed configuration (Fig. 1–4). However, Carr et al. (1998) and Chapko et al. (1991) did not find a total forage yield advantage in different barley–pea and oat–pea intercropping studies using the sole crop seeding rates.
Application of fertilizer N decreased the LER both on biomass and protein bases, especially at a higher N level (Fig. 1–4). Jensen (1996) found that fertilizer N did not influence the total intercrop grain yield, but decreased the proportion of pea contributing to yield. A number of other studies have indicated that the LER of intercrops tend to be higher under low N conditions (Ahmed and Rao, 1982; Martin and Snaydon, 1982; Hiebsch and McCollum, 1987; Ofori and Stern, 1987). This result has practical implications for organic farming where inorganic N fertilizer is not available.
Over-application of fertilizer N may result in elevated nitrate content in harvested hay. For example, nitrate measurements for the forage samples from WARC, showed 6.6 g kg–1 of NO3– content at N2 treatment (134 kg N ha–1) in 2001, which is potentially toxic to cattle (Bos sp.) (Fjell et al., 1991).
The NDF and ADF contents—measures for feed digestibility fiber and feed intake potential—are two other indices used to evaluate forage quality. Brundage and Klebesadel (1969) and Chapko et al. (1991) reported that intercropping cereal with legume showed decreases in NDF and ADF content in the cereal component compared with the sole cropped cereal. Carr et al. (2004) also reported decreased forage NDF concentration when barley was intercropped with pea, but forage ADF concentration was unaffected. In this study, based on one measurement at WARC in 2001, the NDF and ADF concentrations of the combined forage were unaffected by intercrop row arrangements. The ADF concentration in the intercrops was similar to the pure barley (from the middle rows of the 4 x 4 configuration), but the combined forage from the 4 x 4 row configuration had a lower NDF concentration than the pure barley.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSNOTESRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
NOTES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSNOTESRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSNOTESRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  I. Vasilakoglou and K. Dhima Forage Yield and Competition Indices of Berseem Clover Intercropped with Barley Agron. J., November 7, 2008; 100(6): 1749 - 1756. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Chen, P. Miller, F. Muehlbauer, K. Neill, D. Wichman, and K. McPhee Winter Pea and Lentil Response to Seeding Date and Micro- and Macro-Environments Agron. J., October 31, 2006; 98(6): 1655 - 1663. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. R. Szumigalski and R. C. Van Acker Nitrogen Yield and Land Use Efficiency in Annual Sole Crops and Intercrops Agron. J., June 5, 2006; 98(4): 1030 - 1040. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Crop Science | Vadose Zone Journal | |||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
