|
|
Impact |
© 2007 Plant Management Network. Impact of Nitrogen Applications to Wheat on No-tillage Double-crop Soybean K. R. Brye, D. E. Longer, M. L. Cordell, and P. Chen, Department of Crop, Soil, and Environmental Sciences, 115 Plant Science Building, University of Arkansas, Fayetteville 72701; E. E. Gbur, Agricultural Statistics Laboratory, 101 Agricultural Annex, University of Arkansas, Fayetteville 72701; and A. L. Pirani, Department of Crop, Soil, and Environmental Sciences, 115 Plant Science Building, University of Arkansas, Fayetteville 72701 Corresponding author: K. R. Brye. kbrye@uark.edu K. R. Brye, D. E. Longer, M. L. Cordell, P. Chen, E. E. Gbur, and A. L. Pirani. 2007. Impact of nitrogen applications to wheat on no-tillage double-crop soybean. Online. Crop Management doi:10.1094/CM-2007-0219-01-RS. Abstract Due to the popularity of double-crop production systems in the southern United States, particularly soybean [Glycine max (L.) Merrill] following wheat (Triticum aestivum L.), wheat-residue management practices are critical to the successful establishment of double-cropped soybean. The objective of this study was to ascertain the effects of N application to wheat on no-tillage soybean growth and production in a double-crop system. This study was conducted over two cropping seasons at two locations on silt-loam Alfisols in eastern Arkansas. Wheat grain yield and subsequent residue mass generally increased as total applied N increased. Soybean plant population, height, and growth-stage rating approximately 30 days after planting were generally unaffected by total applied N to wheat or wheat-residue level. These results indicate that there is no negative impact to soybean planted without prior tillage into a wide range of wheat-residue levels and demonstrates that technological advancements with field implements have been able to overcome the concern of planting into high-residue situations. Introduction Double-crop production systems have become prevalent in the southern United States (7) due in part to more favorable growing conditions (i.e., temperature and moisture) and increased profit margins from maximizing agricultural use of the land. In particular, double-cropped soybean [Glycine max (L.) Merrill] following winter wheat (Triticum aestivum L.) is popular in southern and mid-southern states. In Arkansas, double-crop soybean following wheat accounted for an average of 25% of the total planted soybean area over the last 20 years (10). However, a successful wheat-soybean double-crop system is contingent on wheat-residue management practices used prior to establishing soybean as the second crop in the rotation. Differential post-harvest crop residue levels are a function soil fertility and production practices. Generally, more fertile soils, whether due to inherent fertility or to fertilization, produce greater grain and biomass yields. Although crop residues serve a number of beneficial functions, including protecting the soil form erosion, water conservation, and maintenance of soil organic matter, high levels of crop residues have traditionally been viewed as a hindrance to agricultural production due to planting difficulties, poor stand establishment, and ultimately yield reductions. Crop residues have been shown to interfere with herbicide applications (12), decrease herbicide efficacy (12), and in some cases stimulate a yield-limiting allelopathic effect (2,6,12). Therefore, in the Mississippi River Delta region, crop residues, particularly wheat residues, are commonly burned or incorporated followed by disking to prepare a seedbed for double-cropped soybean (7,8,12). Residue burning is a labor-saving practice that often results in an increased profit margin from the double-crop system (6,16). However, burning has several adverse environmental and ecological impacts. The combustion of dead plant material adds carbon dioxide to the atmosphere and prevents the return of carbon to the soil. In addition, there are legal and safety issues to consider. Thus there is motivation to explore alternative wheat-residue management practices and evaluate their effect on subsequent soybean growth and production when soybean is no-tilled into wheat residue. Recent technological improvements in field equipment have made planting into high-residue conditions more feasible (8). Consequently, adoption of conservation tillage practices [i.e., ridge-tillage (persistent ridges are maintained from year to year where tillage and planting only occur atop the ridges), mulch-tillage (> 30% residue-covered surface), and no-tillage (100% residue-covered surface)] has also increased. For example, the soybean area planted in a double-crop system using conservation tillage practices has increased by almost 12,350 acres/year and has averaged approximately 32% of the total planted soybean area between 1989 and 2002 in Arkansas (3). More specifically, the soybean area planted as no-tillage in a double-crop system in Arkansas has averaged approximately 20% of the total planted soybean area and has increased by nearly 17,300 acres/year since 1989 (3). Since research documenting a negative relationship between wheat-residue level and soybean yield in a double-crop system (2,5,6,9,11,15) is rather out dated, the effects of common production practices associated with the wheat-soybean double-crop system need to be re-evaluated in the context of improved production practices and field equipment. Therefore, the objective of this study was to ascertain the effects of N fertilization of wheat on no-tillage soybean growth and production in a double-crop system. We hypothesized that soybean growth and production parameters, particularly plant height, population, growth-stage ratings, and yield, are unaffected by N fertilization of wheat. Site Description Research was conducted over two wheat-soybean rotation cycles (in 2001-2002 and 2002-2003) at two locations representing typical wheat-soybean double-cropping regions in the Mississippi River Delta region of eastern Arkansas. The study was conducted on Calhoun silt-loam (fine-silty, mixed, active, thermic, Typic Glossaqualf) at the University of Arkansas Pine Tree Branch Station (Pine Tree; 35°7’10.54”N, 90°45’51.56”W near Colt, AR) and on a Calloway silt loam (fine-silty, mixed, active, thermic Aquic Fraglossudalf) at the University of Arkansas Cotton Branch Experiment Station (Cotton Branch; 34°44’2.26”N, 90°45’51.56”W near Marianna, AR). Experimental Design and Treatments A randomized complete block design with three replications of 10 treatments was established at both locations. Treatments consisted of varying N rates applied to the wheat crop, either as a single application at the early-jointing stage or as a split application with the additional N being applied at the late-jointing stage, to produce a range of wheat residue levels. Single applications of N included 0, 20, 30, 50, 70, and 90 lb/acre, and split applications of N included 90 + 45, 90 + 90, 100 + 100, and 120 + 120 lb/acre. The N source was pelletized urea (46% N) and was broadcast applied by hand to all plots. The standard N fertilization practice for wheat on irrigated silt-loam soils in Arkansas is a split application of N at 90 + 90 lb/acre. Field Management Prior to the initiation of this study, both study locations were cropped under conventional tillage methods; thus the results of this study represent a short-term no-tillage history. Grain sorghum [Sorghum bicolor (L.) Moench] and soybean were previously grown in a non-double-cropped system at Pine Tree and Cotton Branch, respectively. Prior to initial wheat planting, the plot area was disked twice followed by landplaning (i.e., surface smoothing) and field cultivation at Pine Tree and disked twice followed by field cultivating at Cotton Branch. In addition, prior to initial wheat planting at Cotton Branch, a 200-lb/acre broadcast application of 9-23-30 blended fertilizer was applied. In fall each year, a single wheat cultivar (Coker 9663) was drill-seeded with 6-inch row spacing at a rate of 100 lb/acre at both sites in both years. Each following spring, 10- by 20-ft plots were established at both locations; the same plots established in 2002, were used in 2003. The first N application was made in early March and the split application was made in late March each year. Wheat was harvested in early- to mid-June each year. Wheat grain was collected from the middle 5 ft of each plot with a small-plot combine and oven dried for wheat yield determination. Within two weeks after wheat harvest, a single glyphosate-resistant soybean cultivar (Pioneer 95B32, maturity group 5.3) was drill-seeded without tillage with a row spacing of 7.5 inches at a rate of 90 lb/acre at Pine Tree and 42 lb/acre at Cotton Branch in 2002 and 80 lb/acre at Pine Tree and 96 lb/acre at Cotton Branch in 2003, which, assuming an average of 130 mg per seed, corresponds to 314 500, 146 500, 279 500, and 324 500 seeds/acre for Pine Tree and Cotton Branch in 2002 and 2003, respectively. The low soybean planting rate at Cotton Branch in 2002 was an experimental oversight that was not noticed until several weeks after emergence. The final average soybean density was 5 and 1 plants/ft at Pine Tree and Cotton Branch, respectively, in 2002 and 1.8 and 3.5 plants/ft at Pine Tree and Cotton Branch, respectively, in 2003. In addition, there was at least one rainfall event between wheat harvest and soybean planting each year to somewhat equilibrate soil moisture among all plots. Soybeans were furrow-irrigated at Cotton Branch and flood-irrigated at Pine Tree three times throughout each soybean growing season following 7 to 10 days without significant rainfall. Due to prolonged wet field conditions, two adjacent 3-ft row sections of total above-ground biomass were collected by hand from each soybean plot from both locations in early November 2002. Soybean biomass samples were oven dried at 55°C for 7 days and mechanically thrashed to remove and collect all seeds. In 2003, a plot combine was used to harvest the middle 5- by 20-ft section of each plot at both locations in late October 2003. All grain from the plot combine harvest was collected for total plot yield determination on an oven-dry basis. Soil Sampling and Analyses Prior to soybean planting in both years, two sets of soil samples were collected. One set consisted of a single 2-inch diameter soil core collected from the 0- to 4-inch depth using a slide hammer for soil bulk density and particle-size distribution determinations. The second set of soil samples consisted of ten 1-inch diameter cores randomly collected and composited from the top 4 inches of each plot for soil chemical analyses. Soil samples were oven dried at 70°C for 48 h, crushed, and sieved through a 0.08-inch (2-mm) mesh screen. Fifty-gram subsamples of oven-dry soil were used for particle-size analysis using the hydrometer method (1). Dried and sieved soil was extracted with Mehlich-3 extractant solution (14) in a 1:10 soil-to-extractant-solution ratio (w/v) and analyzed for extractable phosphorus (P) and potassium (K) using an inductively coupled argon-plasma spectrophotometer (CIROS CCD model, Spectro Analytical Instruments, MA). Inorganic soil nitrate-N (NO3-N) and ammonium-N (NH4-N) were determined by the standard 2N potassium chloride (KCl) extraction followed by colorimetric analysis on an autoanalyzer. Extractable soil P and K and inorganic soil N concentrations were multiplied by the corresponding measured bulk density from each plot to express soil nutrient contents on a mass-per-area basis (i.e., lb/acre). Soil pH and electrical conductivity (EC) were determined with an electrode on a 1:2 soil-to-water solution (w/v). Organic matter was determined by weight-loss-on-ignition after 2 h at 360°C (13). Residue Sampling Following wheat harvest, but prior to soybean planting each year, the residue that had passed through the combine was uniformly spread by hand back onto the plot from which it came and all residue was cut to the soil surface with a tractor-mounted rear rotary mower. Mowing is a practice not commonly conducted in eastern Arkansas, but was done to create a residue-covered surface that resembles residues produced after wheat harvests under no-tillage conditions. Wheat residue mass was determined by collecting all plant material (i.e., wheat residue plus weeds) within a 19.6- by 19.6-inch (0.25 m²) metal frame place randomly within each plot immediately following wheat harvest and residue chopping, but prior to soybean planting. Residue samples were dried for 5 days at 55°C for dry matter determination. Soybean Growth and Production Parameters Soybean population estimates were obtained between 30 and 33 days after planting (DAP) in both years by averaging the number of soybean plants within two 3-ft sections of row in opposite corners of the plots. Vegetative growth stages were determined between 30 and 33 DAP both years using a soybean growth staging system (4), which is based on the number of fully developed trifoliates above the first node. Plant heights were also measured between 30 and 33 DAP in both years. Oven-dry soybean seed yield was determined from hand harvesting in 2002 and combine harvesting in 2003. Statistical Analyses An analysis of variance (ANOVA) was used to test the uniformity of soil particle-size fractions, soil bulk density, and soil chemical properties among plots with differing N rates applied to wheat at each location. Regression analyses were conducted separately by year and location to ascertain the relationships between total applied N and/or wheat-residue level and inorganic soil N and soybean growth (i.e., plant height, population, and growth-stage rating) and production (i.e., yield) parameters. Regression analyses assumed both total applied N and wheat-residue level in the two separate analyses represented continuous independent variables. Year and location were not explicitly tested as factors due to dissimilar cropping histories between locations, dissimilar fertilization schemes prior to the initial wheat crop, and dissimilar soybean seeding rates between locations and years; thus, results are presented based on four location-year combinations. All statistical analyses were conducted using Minitab Version 13.31 (Minitab Inc., State College, PA). Wheat Response to N Fertilization As expected, wheat grain yield increased linearly (P ≤ 0.015) as total applied N increased at both locations and in both years (Fig. 1). Since the objective of this study was not to predict wheat yield nor to identify the total applied N rate at which wheat yield failed to respond to N fertilization, linear regression was used for all data. The effects of split versus single applications of N on wheat yield were beyond the scope of this study and, therefore, not evaluated. As would also be expected, based on the significant N fertilization effect and assuming a similar harvest index among all treatments, total above-ground dry matter production also increased as total applied N increased at both locations and in both years (P ≤ 0.05; data not shown). Wheat-residue mass was unaffected by total applied N at both locations in 2002 (Fig. 2). However, wheat-residue mass into which the subsequent soybean crop was planted increased linearly as total applied N increased at both locations in 2003 (P < 0.05; Fig. 2). Total above-ground vegetative wheat biomass (i.e., total biomass minus grain) was unrelated (P > 0.5) to the actual amount of wheat residue remaining on the soil surface following wheat harvest and mowing, but prior to no-tillage soybean planting.
Wheat-residue levels, averaged across all N rates, were similar between locations in 2002, averaging 2535 (SE = 137) lb of residue per acre. However, in 2003, the soybean crop at Cotton Branch was planted into significantly more (P < 0.001) surface residue than at Pine Tree, where the mean wheat-residue mass was 3856 (SE = 205) lb of residue per acre at Cotton Branch and 1707 (SE = 99) lb of residue per acre at Pine Tree. The nearly two-fold difference in wheat-residue levels between locations (Fig. 2) in 2003 indicates that parameters other than applied N had greater effects on residue production.
Soil Properties Prior to Soybean Planting Sand, silt, and clay percentages and soil bulk density in the top 4 inches did not vary due to N treatment at either location in 2002 (Table 1). Similar to soil physical properties, Mehlich-3 extractable soil P and K contents, organic matter, soil pH, and EC in the top 4 inches did not vary among treatments at either location or in either year (Table 1), except for extractable P at Cotton Branch in 2003 where, on average, extractable P was higher in the 0 lb of N per acre treatment than in the 90 lb of N per acre treatment. Despite few differences in soil fertility among plots at either location, N fertilization of wheat likely affected subsequent no-tillage, double-crop soybean growth and production. Table 1. Summary of particle-size distributions and selected soil chemical properties in the top 4 inches for no-tillage soybean grown at two locations for two years in eastern AR.
x Mean (± standard error); n = 30. y EC = Electrical conductivity. It is commonly known that most legume crops will use inorganic N present in the soil before fixing their own N. Though soybean is leguminous and therefore has the ability to fix its own N, residual inorganic soil N that was not extracted by the previous wheat crop could have affected early growth of the subsequent soybean crop and likely to a greater degree than the few P differences among plots at Cotton Branch in 2003. However, neither NO3-N, NH4-N, nor total inorganic N (NO3-N + NH4-N) differed among plots at Cotton Branch in either year or Pine Tree in 2002 when sampled following wheat, but prior to soybean (data not shown). At Pine Tree in 2003, residual NO3-N, and consequently total inorganic N, increased (P < 0.001) as total applied N increased (Fig. 3).
Soybean Response to N Fertilization of Wheat Early-season soybean growth parameters were generally unresponsive to total applied N to the previous wheat crop. Soybean populations approximately one month after planting were unaffected by total applied N at either location and in either year. Soybean height was unaffected by total applied N in three of four year-location combinations. Soybean growth-stage ratings were unaffected by total applied N in two of four year-location combinations. Soybean plant height (P = 0.006) and growth-stage rating (P = 0.003) increased as total applied N increased at Cotton Branch in 2002 (Fig. 4) despite no relationship between total applied N and inorganic soil N, nitrate-N, ammonium-N, or the sum of both, in 2002. The lack of a correlation between total applied N and inorganic soil N at wheat harvest in 2002 indicated that all applied N was depleted from the root zone and that there was no carry-over N to affect subsequent early-season soybean growth.
Soybean growth-stage rating tended to increase (P = 0.016), meaning the plants were at a more-advanced growth stage, as total applied N increased at Pine Tree, but not at Cotton Branch, in 2003 (Fig. 5). This result has several plausible, but inseparable, explanations. Additional inorganic soil N in the root zone at the time of soybean planting (i.e., carry-over N) (Fig. 3) may have contributed to improved soybean stand establishment and more rapid advancement of soybean growth. Similarly, the positive correlation between wheat-residue mass and total applied N in 2003 may have stimulated an early-season soybean response to residue level.
With one exception, neither soybean population, height, nor growth-stage rating were related to wheat-residue level at either location in either year. Despite the lack of wheat residue level and soybean height response to total applied N at either location in 2002 (Fig. 2), soybean height increased significantly as wheat-residue level increased (P = 0.011) at Pine Tree in 2002 (Fig. 6). Therefore, factors other than total applied N and residue level were likely responsible for the significant height and growth stage response to total applied N at Cotton Branch and height response to residue level at Pine Tree in 2002.
The association of taller plants with higher wheat-residue levels may be related to two phenomena. Although not specifically measured, in theory, slightly higher soil moisture levels associated with the greater evaporation barrier of a thicker, more massive residue cover may have stimulated additional vegetative growth, especially during early growth stages. A more likely explanation would involve a phototropic response of stimulated stem elongation due to low-light conditions from under a thicker, more massive residue cover. This elongation and slight increase in plant height at Pine Tree in 2002 was a residue-induced effect that was of little consequence. Soybean yield was unaffected by total applied N to wheat or wheat-residue level in either year and at either location (Fig. 7); thus indicating that soybean yield was not negatively impacted across residue levels. There was no evidence of lodging increase across residue levels and, in general, lodging was minimal throughout the plots at both locations and in both years.
The lack of soybean yield response to wheat-residue level in this study is consistent to that previously observed by Caviness et al. (2) for no-tillage, double-crop soybean in eastern Arkansas. However, these results are in contrast to that of Vyn et al. (15) who documented a significant negative correlation (r > 0.89, P < 0.01) between yield and surface residue level at two locations over three years in southwestern Ontario, Canada where the climate is much cooler than that of the mid-Southern United States. Practical Implications As technological advancements are made with field equipment, previous limitations, such as planting into high-residue conditions, are overcome, allowing alternative residue management practices to be explored. This is especially important for the wheat-soybean double-crop production system common to the mid-Southern and Southern United States where the environmentally unfriendly practice of residue burning is prevalent. This study demonstrated that soybean growth and production are not negatively influenced by increasing wheat-residue levels. Results of this study suggest that burning, as a residue management practice to reduce the potential negative impact of planting into a high-residue situation, may be unnecessary. The technology available with today’s field implements has largely overcome the planting problems due to the presence of a large amount of surface residue. Further research is needed to evaluate (i) the agronomic viability of wheat-residue management alternatives to the combination of burning followed by conventional tillage and (ii) the long-term impact of increased wheat residue on soil organic matter, fertility, and soybean production in the wheat-soybean double-crop production system. Acknowledgments Funding for this project was provided by the Arkansas Soil Testing and Research Board. William Johnson, Trey Reaper, Claude Kennedy, Roger Eason, Shawn Clarke, and other research station personnel are gratefully acknowledged for their support in conducting this project. Literature Cited 1. Arshad, M. A., Lowery, B., and Grossman, B. 1996. Physical tests for monitoring soil quality. Pages 123-141 in: Methods for Assessing Soil Quality. J. W. Doran and A. J. Jones, eds. Soil Sci. Soc. Am. Spec. Publ. 49. Madison, WI. 2. Caviness, C. E., Collins, F. C., and Sullivan, M. 1986. Effect of wheat residue on early growth of soybean. Ark. Farm Res. 35:8. 4. Fehr, W. R., Caviness, C. E., Burmood, D. T., and Pennington, J. S. 1971. Stage development descriptions for soybeans, Glycine max (L.) Merrill. Crop Sci. 11:929-931. 5. Guy, S. O., and Oplinger, E. S. 1989. Soybean cultivar performance as influenced by tillage system and seed treatment. J. Prod. Agric. 2:57-62. 6. Hairston, J. E., Sanford, J. O., Pope, D. F., and Horneck, D. A. 1987. Soybean-wheat doublecropping: Implications from straw management and supplemental nitrogen. Agron. J. 79:281-286. 7. Heatherly, L. G., Elmore, C. D., Wesley, R. A., and Spurlock, S. R. 1996. Yield and net returns from monocrop winter wheat systems and a winter wheat-soybean doublecrop system. J. Prod. Agric. 9:61-65. 8. Kelley, K. W., and Sweeney, D. W. 1998. Effects of wheat-residue management on double-cropped soybean and subsequent crops. J. Prod. Agric. 11:452-456. 9. Kladivko, E.J., Griffith, D. R., and Mannering, J. V. 1986. Conservation tillage effects on soil properties and yield of corn and soya beans in Indiana. Soil Till. Res. 8:277-287. 11. Roland, C. E. 1992. Crop sequence and tillage system effects on soybean performance and soil physical properties. M.Sc. thesis, Univ. of Guelph, Guelph, Ontario, Canada. 12. Sanford, J. O. 1982. Straw and tillage management practices in soybean-wheat double-cropping. Agron. J. 74:1032-1035. 13. Schulte, E. E., and Hopkins, B. G. 1996. Estimation of soil organic matter by weight loss-on-ignition. Pages 21-31 in: Soil Organic Matter: Analysis and Interpretation. F. R. Magdoff, M. A. Tabatabai, and E. A. Hanlon, eds. Soil Sci. Soc. Am., Madison, WI. 14. Tucker, M. R. 1992. Determination of phosphorous by Mehlich 3 extraction. Page 6-8 in: Soil and Media Diagnostic Procedures for the Southern Region of the United States. S. J. Donohue, ed. Vir. Agric. Exp. Stn. Ser. Bull. 374. Blacksburg, VA. 15. Vyn, T. J., Opoku, G., and Swanton, C. J. 1998. Residue management and minimum tillage systems for soybean following wheat. Agron. J. 90:131-138. 16. Wesley, R. A., and Cooke, F. T. 1988. Wheat-soybean double-crop systems on clay soil in the Mississippi Valley area. J. Prod. Agric. 1:199-171. |
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
