| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Southern Research and Outreach Center, Univ. of Minnesota, 35838 120th St., Waseca, MN 56093-4521
b Dep. of Soil, Water and Climate, 1991 Upper Buford Cir., Univ. of Minnesota, Saint Paul, MN 55108
* Corresponding author (jvetsch{at}umn.edu)
Received for publication May 12, 2006.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: CP, chisel plow • GWC, gravimetric water content • NT, no-till • PR, penetrometer resistance • RC, row cultivation • SFC, spring field cultivate • ST, strip-tillage • ZT, zone-tillage
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
The quantity of available literature on rotational tillage systems describing residue cover, yield, and economic return is not extensive. Some studies have focused on periodic full-width moldboard plow or CP tillage (Dickey et al., 1983; Randall et al., 1996); while others have evaluated deep ZT or subsoiling implements and their immediate and residual effects (Varsa et al., 1997; Soane et al., 1987; Marks and Soane, 1987; Vetsch and Randall, 2002). Generally, rotational tillage, either full-width (moldboard plow and CP) or deep ZT, has improved crop yields on moderately well to somewhat poorly drained soils (Dickey et al., 1983; Randall et al., 1996) and/or soils with dense layers (Varsa et al., 1997). Other researchers have found little to no effect of tillage on soils varying from poorly to well drained (Soane et al., 1987; Marks and Soane, 1987; Vetsch and Randall, 2002; Vepraskas et al., 1995). Randall et al. (1996) reported long-term NT reduced corn yields compared with conventional tillage, thus resulting in significantly greater risk to growers on poorly drained glacial till soils of south-central Minnesota. Yield reductions were magnified by unusual growing season climatic conditions (generally worse in wet and dry years). Vetsch and Randall (2002) found no corn yield response to semiannual, 38-cm-deep ZT in rotation with NT soybean on a silt loam in Minnesota.
No-till soybean production throughout the Corn Belt is more commonplace than NT corn production. The effects of tillage on soybean production are covered extensively in the literature. Yin and Al-Kaisi (2004) compared soybean yield in various tillage systems at five Iowa sites for periods ranging from 8 to 15 yr. At four of five sites (one site did not include CP tillage) NT soybean yields were not different compared with CP, when averaged across years. No-tillage reduced soybean yields compared with CP in one of 15 yr at the Nashua site and two of 6 yr at the Newell site. Significant differences between NT and CP were not found at the Crawfordsville site. When NT soybean yields were compared with other tillage systems over time (5-yr intervals), they found neither significant improvement nor deterioration of yield. They concluded NT soybean yields were usually within 5% of other tillage systems, but generally had equal or greater economic returns. Lueschen et al. (1992) found soybean yields were not affected by tillage on one of three glacial till sites in Minnesota, whereas significant yield differences were found in six of 11 site-years at two other sites. Yield differences among tillage treatments (moldboard, CP, spring disk, NT, ridge-till, and till-plant) at these two sites were small, ranging from 0 to 0.2 Mg ha–1 and were not consistent between sites. The authors concluded that the inconsistency in soybean yields among tillage systems makes it nearly impossible to single out any one tillage system as best for soybeans after corn. Randall et al. (1996) measured the effects of rotational tillage on soybean yields in a corn and soybean rotation. They found NT soybean yields in latter years of the study period were 0.3 to 0.4 Mg ha–1 less than when rotational full-width (CP and moldboard plow) tillage was used for corn. They suggested surface soil compaction and less-than-ideal weed control in some years may have led to these yield reductions. Randall et al. (2001) found soybean yields were not affected by tillage treatments [NT, one-pass (spring disk), and CP+SFC] on a low-P testing site in Minnesota, while on a high-P testing site NT yields were 0.1 to 0.2 Mg ha–1 less than the one-pass disk and CP+SFC tillage treatments. In Illinois on silt loam soils with claypans (southern locations) or dark–colored prairie soils (northern locations), Nafziger (2003) found soybean yields were not affected by deep tillage, although improvements in surface soil compaction and permeability were noted.
Many researchers have used soil penetrometer resistance (PR, cone index) to quantify benefits of periodic tillage in NT systems (Bauder et al., 1981; Busscher et al., 1986; Pierce et al., 1992; Vetsch and Randall, 2002). Bauder et al. (1981) showed long-term NT (10 yr) had greater PR than moldboard or CP on a clay loam soil. Busscher et al. (1986) found residual effects of subsoiling could be seen in the year following subsoiling on a coastal plain soil. However, PR in the year-old subsoil trench was considered to be root-limiting (1.5–2.5 MPa). They also noted that CP did not disturb the soil to the desired depth and took more energy than subsoiling. Pierce et al. (1992) found that PR approached 3.0 MPa in NT on a well-drained loam soil in Michigan but was reduced to 0.5 MPa by ZT and remained <1.0 MPa during the next 2 yr. Vetsch and Randall (2002) found in-row PR ranged from 0.2 to 1.1 MPa on a silt loam soil. Generally, ST (20-cm depth) and deep ZT (38-cm depth) reduced PR nearly to the depth of tillage compared with NT and one-pass, SFC (10-cm depth). However, PR between 45 and 60 cm was not different among treatments.
It is evident additional research is needed to better understand yield reductions in long-term NT and yield responses of corn and soybean to rotational tillage systems, especially on the highly productive but poorly drained soils of the northern Corn Belt. The primary objective of this study was to quantify the effects of rotational full-width tillage compared with long-term NT and ZT (20- and 38-cm-deep) systems with and without in-season RC on corn and soybean yield and economic return on a tile-drained clay loam soil in south-central Minnesota. A secondary objective was to characterize the effects of these tillage systems on residue coverage, early growth of corn, and soil PR.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Eighteen treatments were arranged in a randomized, complete-block design with four replications (blocks) for corn and an adjacent four replications for the soybean experiment. Sixteen of the 18 treatments comprised a factorial arrangement of three factors: tillage treatment for corn following soybean (NT, ZT, ST, and SFC), tillage treatment for soybean following corn (NT and CP+SFC), and in-season RC for corn (with or without). The other two treatments were annual CP+SFC treatments for both crops—one with RC for corn the other without RC. Each individual plot was 3 m wide (four 76-cm rows of corn) by 17.5 m long.
All corn plots, regardless of tillage treatment for corn, were planted with a John Deere (Moline, IL) Model 7300 planter equipped with Yetter (Colchester, IL) Residue Managers and a 36-cm-diam. ripple coulter mounted 5 cm to the side of the row. An injection nozzle located behind this coulter was used to inject a stream of liquid starter fertilizer to an 8-cm depth for all corn plots. Starter N and P were applied at 12 and 18 kg ha–1, respectively, in 2000, 2002, and 2003, and at 24 and 36 kg ha–1, respectively, in 2001. Corn [Pioneer hybrid 37H26 (37H27 in 2002)] was planted on 14 May 2001, 3 May 2002, and 29 Apr. 2003 at 79 000 seeds ha–1. Because a late spring frost on 17 May 2000 reduced stand, the study was replanted on 23 May 2000.
The NT treatment received no fall or preplant tillage before planting corn. The ZT treatment was deep zone-tilled to a 38-cm depth on 76-cm centers in the fall with a four-shank John Deere (Moline, IL) Model 955 Ripper (Vetsch and Randall, 2002). The ST treatment was performed to a 20-cm depth in the fall with a DMI (CASE DMI, Goodfield, IL) ST unit with mole knives (Randall and Hill, 2000). Corn was planted the following spring directly into the ST and ZT zones without preplant tillage. The SFC treatment was field cultivated to a 7- to 10-cm depth before planting. The CP+SFC treatment was chisel plowed to a 20-cm depth in the fall with a Hiniker (Mankato, MN) CP and field cultivated just before planting. In mid to late June, one-half of the treatments received RC to an 8-cm depth with a John Deere (Moline, IL) Model RM cultivator.
Nitrogen as urea with NBPT [N-(n-butyl)thiophosphoric triamide] was broadcast applied at 146 kg ha–1 on 9 May 2000. In 2001, 2002, and 2003, N was injected 10-cm deep between the rows as 28% UAN (urea–ammonium nitrate) at 146 kg N ha–1 as an early sidedress application. Weed control was performed with a combination of pre- and postemergence herbicide applications. Weed control was excellent in all years of the study.
Surface residue measurements were taken at a 45° angle to the rows after planting using the line-transect method described by Sloneker and Moldenhauer (1977). Extended-leaf plant heights were taken from 10 random plants 
35 d after emergence. Corn grain and soybean seed yield and moisture content were measured by harvesting two 17-m-long rows with a plot combine. Corn grain and soybean seed yields were expressed on a 15.5 and 13% moisture basis, respectively.
Soybean was planted in six 51-cm rows with a Hiniker (Mankato, MN) planter equipped with Dawn (Dawn Equipment, Sycamore, IL) row cleaners. The 51-cm row spacing of the planter resulted in soybean rows that were 13 or 38 cm from the previous year's corn rows in the NT treatments. Soybeans (Pioneer 91B64) were seeded at 325000 seeds ha–1. Inadequate stand establishment in 2002 required a replant on 30 May. In other years soybeans were planted on 1 May 2000 and 17 May 2001. The NT treatment received no fall or preplant tillage before planting soybean and the corn stalks were not chopped or shredded. The CP+SFC treatment was chisel plowed to a 20-cm depth in the fall with a Hiniker CP and field cultivated (8–10 cm deep) just before planting. Excellent weed control was obtained with postemergence applications of glyphosate herbicide.
Net economic return of corn and soybean production was calculated based on tillage cost estimates (Lazarus and Selley, 2005), corn and soybean yields from 2000 through 2002, and $29 and $73 Mg–1 prices (USDA "county loan rate") of corn and soybean, respectively. The gross income from corn and soybean production was summed for each plot before subtracting all of the tillage treatment costs for the corn–soybean rotation. The net return was then divided by two, assuming an equal proportion of corn and soybean in the crop rotation.
Penetrometer resistance was measured in 15-mm depth increments directly in the row area for all four tillage treatments with a Rimik CP10 recording penetrometer (Agridry-Rimik Pty. Ltd., Toowoomba, QLD, Australia) with a standard ASAE 30°, 12.8-mm cone. Three measurements from each main plot were taken in mid-April in 2000 (before spring tillage), mid-May in 2000 through 2002 (at emergence), and mid-June in 2000 [about V4 (Ritchie et al., 1986)]. At the same time and from the same area as PR measurements, soil samples (four cores per plot) were collected and analyzed for gravimetric water content (GWC) in increments of 0–10, 10–20, 20–30, and 30–45 cm. The in-row area in NT, SFC, and CP+SFC treatments was approximated for the April sampling. Statistical analysis and graphical display of PR data were simplified by combining the 15-mm PR measurements into a 75-mm increment and then averaging.
Daily precipitation and air temperature for the growing season were recorded at a weather station 0.3 km from the site. Analysis of variance statistics were performed using the GLM procedure of SAS (SAS Inst., 1999). All LSDs were calculated at P 
0.05 for residue cover and crop production parameters and P 0.10 for soil GWC and PR.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
|
|
Soybean Seed Yield
Soybean seed yield was affected by the main effects of year and tillage for soybean, but not by tillage or RC for corn (Table 2). Soybean yields were 0.07 Mg ha–1 (1 bu ac–1) greater with CP+SFC tillage for soybean compared with NT when averaged across years, tillage for corn, and RC for corn. This small yield difference was similar to those found by Lueschen et al. (1992).
Several significant interactions were found for soybean yield including: year x tillage for soybean, tillage for corn x tillage for soybean, RC x tillage for soybean, and tillage for corn x RC (Table 2). These findings may help explain soybean yield responses in reduced and rotational tillage systems on poorly drained clay loam soils. The year x tillage for soybean interaction indicated the yield effects of tillage for soybean varied among years (data not shown). Soybean yields with NT were 0.24 Mg ha–1 less than CP+SFC tillage in 2000, but not significantly different in 2001 and 2002. Reduced yields with NT in 2000 may have been related to a wetter-than-normal spring. The tillage for corn x tillage for soybean interaction (data not shown) is explained by soybean yields being increased 0.12 Mg ha–1 or more when CP+SFC tillage was used for soybean compared with NT in NT and ZT systems for corn, but not in SFC or ST systems for corn. When CP+SFC tillage for soybean followed ZT for corn, soybean yield was greater than any other treatment combination, suggesting a yield benefit for rotational full-width tillage. However, the yield response obtained with CP+SFC tillage was negated when RC was used for the previous corn crop. Under these conditions periodic shallow tillage, like RC for a previous corn crop, was apparently adequate to disrupt residue accumulation and soil consolidation and enhance soybean growth. This treatment combination (NT for soybean following corn that received RC) had yields similar to CP+SFC tillage for soybean, while maintaining about 78% residue cover (data not shown). The tillage for corn x RC interaction resulted from a soybean yield response when RC was coupled with NT for corn, but no response when RC for corn was used with other tillage systems for corn (data not shown). These data show that some form of rotational tillage, however minimal for surface soil disturbance, was needed to maximize soybean yield.
The yield responses to surface soil disturbance found in this study may be related to the placement of the soybean rows with respect to the previous year's corn rows. The 51-cm row spacing of the soybean planter resulted in four soybean rows that were 13 cm from a previous year's corn row and two soybean rows that were 38 cm or midway between the previous year's corn rows in the NT for soybean treatments. Moreover, when NT was used for soybean, these two 38-cm rows were seeded directly into wheel traffic areas from the previous year's corn crop. Slightly slower emergence and early growth were observed in these rows, especially following NT for corn without RC, indicating a less-than-ideal seed bed under these undisturbed conditions. However, these findings are not unique to the row spacing used in this study. Any corn–soybean rotation, where the row spacing for soybean is less than used for corn, would experience NT soybean rows planted into wheel-tracked areas from the previous crop.
Economic Return to Tillage Costs
Economic return for the combined corn–soybean rotation was affected by the main effects of year and RC (Table 2). Economic return, averaged across tillage treatments, was significantly greater in 2002 compared with other years because of greater corn yields. Row cultivation of corn reduced economic return $7 ha–1 compared with no RC, when averaged across year, tillage for corn, and tillage for soybean. The small corn and soybean yield responses to tillage for soybean and corn yield response to tillage for corn found in this study did not affect the economic return of those main effects in this study.
Significant interactions were found for year x tillage for soybean, year x tillage for corn x tillage for soybean, and RC x tillage for soybean. The year x tillage for soybean interaction showed CP+SFC tillage for soybean increased economic return in 2000 and 2002, but reduced economic return in 2001 (data not shown). These data parallel the differences found with corn yield, which suggests CP+SFC tillage for soybean had a greater effect on corn yield than it did on soybean yield. The significant year x tillage for corn x tillage for soybean interaction (data not shown) was similar to the year x tillage for soybean interaction. In 2001, NT for soybean had greater economic return than CP+SFC tillage for soybean for all tillage for corn systems, whereas in 2000 and 2002 CP+SFC tillage for soybean had significantly greater return than NT for soybean but only when ZT or ST was used for corn. Moreover, in 2000 and 2002, greatest economic returns were obtained when ZT or ST was used for corn with CP+SFC tillage for soybean (data not shown). The RC x tillage for soybean interaction was illustrated by RC of the previous year's corn, increasing economic return $9 ha–1 when NT was used for soybean, whereas RC of corn reduced soybean return $23 ha–1 when CP+SFC tillage was used for soybean (data not shown). In summary, rotational tillage practices are likely to maximize yields of corn and soybean but not consistently increase economic return on these clay loam soils.
Surface Residue Cover after Planting Corn
Residue cover measured after planting was affected by all of the main effects (Table 4). Residue cover was 67, 41, 56, and 41% with NT, ZT, ST, and SFC tillage for corn, respectively, when averaged across year, tillage for soybean, and RC. No-tillage for the previous year's soybean crop resulted in an 11 percentage point greater residue cover for corn compared with CP+SFC tillage when averaged across year, tillage for corn, and RC. These percentage residue covers are typical for continuous reduced tillage systems in a corn–soybean rotation in the Northern U.S. latitudes due to slow decomposition. Residue from both the previous soybean crop as well as the prior corn crop is plentiful in the spring.
|
Residue Cover after Planting Soybean
Residue cover, measured after planting soybean, was affected by the main effects of year and tillage for soybean but not by the residual effects of tillage or in-season RC of corn (Table 4). No-tillage for soybean had significantly greater residue cover (79%) compared with CP+SFC (33%). No significant interactions were found. Residue cover in all treatments (data not shown) was greater than the 30% threshold recommended for minimizing soil erosion, when averaged across the 3-yr study period.
Corn Plant Height
Early growth of corn, often measured in terms of plant height, is frequently used by farmers and ag advisors to evaluate the relative performance of conservation tillage systems. Plant height 35 d after emergence was affected by all of the main effects (Table 4). When NT was used for soybean compared with CP+SFC tillage, plant heights were reduced considerably, when averaged across year, tillage for corn, and RC. Plant heights were affected by tillage for corn and were ranked ZT > ST > SFC = NT, when averaged across year, tillage for soybean, and RC. Zone tillage increased plant heights 4% compared with ST, but had no affect on corn yields. Row cultivation of corn decreased plant height although the difference was small and inconsequential. Significant interactions were found for tillage for corn x tillage for soybean, RC x tillage for soybean, and year x tillage for corn. The tillage for corn x tillage for soybean interaction showed plant heights responded more to CP+SFC tillage for the previous year's soybean in SFC and NT for corn compared with ZT and ST for corn (data not shown). These data demonstrate the beneficial effects of in-row tillage and residue removal in the seed–row zone with ST and ZT for greater early growth of corn. The RC x tillage for soybean interaction had no plausible explanation (data not shown).
Soil Gravimetric Water Content
Tillage treatments affected GWC (data not shown) in only five of 48 sampling time/depth increment comparisons: April (0–10 cm) and June (11–20 cm) of 2000, 2001 (0–10 and 31–45 cm), and 2002 (11–20 cm). However, differences were relatively small among tillage treatments within a sampling time/depth increment. Gravimetric water content ranged from 0.255 to 0.281, 0.242 to 0.279, 0.267 to 0.288, 0.227 to 0.250, and 0.274 to 0.297 g g–1 in April (2000), June (2000), 2001 (0–10 cm), 2001 (31–45 cm), and 2002 (11–20 cm), respectively. A factorial analysis of the GWC data from 2000 (tillage for corn x tillage for soybean) showed GWC was not affected by tillage for soybean during any of the sampling times or depths. Variation in soil water content can influence or mask differences in PR. However, it's unlikely the small differences in GWC data influenced the PR measurements in this study.
Penetrometer Resistance
In April (preplant), May (corn emergence), and June (about V4) of 2000, in-row PR measurements were taken to evaluate the immediate and residual effects of fall and spring tillage for corn (Fig. 1
). Penetrometer resistance ranged from 0.25 to 1.43 MPa across sampling dates and tillage treatments. No-till and SFC treatments had similar PR throughout the 60-cm profile in all 3 mo. In April, PR was reduced to a 30-cm depth by ZT and ST and to a 15-cm depth by CP+SFC for corn, compared with NT and SFC. By May, some reconsolidation had occurred in ZT, ST, and CP+SFC for corn, but PR in these treatments was less than NT and SFC at similar depths as in April. In June, PR was reduced by ZT and ST in the 8- to 30- and 8- to 22-cm increments, respectively, compared with NT, SFC, and CP+SFC. Differences among treatments in the 0- to 8-cm increment were small. Like Pierce et al. (1992) and Vetsch and Randall (2002), in-row PR was reduced considerably down to the effective depth of tillage with ZT and ST, but some reconsolidation occurred in May and June. Similar to Dickey et al. (1983), Diaz-Zorita et al. (2004), and Vetsch and Randall (2002), PR never exceeded threshold values where root restrictions would be expected. A factorial analysis of these data found CP+SFC tillage for the previous year's soybean crop reduced PR in the 0- to 15-cm increment compared with NT in April, but not in May or June (data not shown). A significant tillage for corn x tillage for soybean interaction was explained by PR in the NT and SFC systems for corn being less when the previous crop of soybean was CP+SFC compared with NT, but tillage for soybean had no effect on PR with ZT and ST. The immediate effects of tillage for corn on in-row PR were evident at the April and May sampling times for ZT, ST, and CP+SFC, while the residual effects of fall ZT and ST systems for corn were still evident in June.
|
|
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  W. J. Cox, J. H. Cherney, and J. H. Hanchar Zone Tillage Depth Affects Yield and Economics of Corn Silage Production Agron. J., September 1, 2009; 101(5): 1093 - 1098. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. W. Archer and D. C. Reicosky Economic Performance of Alternative Tillage Systems in the Northern Corn Belt Agron. J., March 4, 2009; 101(2): 296 - 304. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. L. Jordan, J. S. Barnes, T. Corbett, C. R. Bogle, P. D. Johnson, B. B. Shew, S. R. Koenning, W. Ye, and R. L. Brandenburg Crop Response to Rotation and Tillage in Peanut-Based Cropping Systems Agron. J., October 21, 2008; 100(6): 1580 - 1586. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Mazzoncini, C. Di Bene, A. Coli, D. Antichi, M. Petri, and E. Bonari Rainfed Wheat and Soybean Productivity in a Long-Term Tillage Experiment in Central Italy Agron. J., September 8, 2008; 100(5): 1418 - 1429. [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 | |||
