| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
USDA-ARS Sustainable Agricultural Systems Lab., Beltsville, MD 20705
* Corresponding author (john.teasdale{at}ars.usda.gov)
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Abbreviations: CC, cover crop system • CV, crownvetch living mulch system • NT, no-tillage system • OR, organic system • SADP, Sustainable Agriculture Demonstration Project
Received for publication December 21, 2006.
USDA-ARS Sustainable Agricultural Systems Lab., Beltsville, MD 20705
* Corresponding author (john.teasdale{at}ars.usda.gov)
Received for publication December 21, 2006.
There have been few comparisons of the performance of no-tillage cropping systems vs. organic farming systems, particularly on erodible, droughty soils where reduced-tillage systems are recommended. In particular, there is skepticism whether organic farming can improve soils as well as conventional no-tillage systems because of the requirement for tillage associated with many organic farming operations. A 9-yr comparison of selected minimum-tillage strategies for grain production of corn (Zea mays L.), soybean [Glycine max (L.) Merr.], and wheat (Triticum aestivum L.) was conducted on a sloping, droughty site in Beltsville, MD, from 1994 to 2002. Four systems were compared: (i) a standard mid-Atlantic no-tillage system (NT) with recommended herbicide and N inputs, (ii) a cover crop-based no-tillage system (CC) including hairy vetch (Vicia villosa Roth) before corn, and rye (Secale cereale L.) before soybean, with reduced herbicide and N inputs, (iii) a no-tillage crownvetch (Coronilla varia L.) living mulch system (CV) with recommended herbicide and N inputs, and (iv) a chisel-plow based organic system (OR) with cover crops and manure for nutrients and postplanting cultivation for weed control. After 9 yr, competition with corn by weeds in OR and by the crownvetch living mulch in CV was unacceptable, particularly in dry years. On average, corn yields were 28 and 12% lower in OR and CV, respectively, than in the standard NT, whereas corn yields in CC and NT were similar. Despite the use of tillage, soil combustible C and N concentrations were higher at all depth intervals to 30 cm in OR compared with that in all other systems. A uniformity trial was conducted from 2003 to 2005 with corn grown according to the NT system on all plots. Yield of corn grown on plots with a 9-yr history of OR and CV were 18 and 19% higher, respectively, than those with a history of NT whereas there was no difference between corn yield of plots with a history of NT and CC. Three tests of N availability (corn yield loss in subplots with no N applied in 2003–2005, presidedress soil nitrate test, and corn ear leaf N) all confirmed that there was more N available to corn in OR and CV than in NT. These results suggest that OR can provide greater long-term soil benefits than conventional NT, despite the use of tillage in OR. However, these benefits may not be realized because of difficulty controlling weeds in OR.
Abbreviations: CC, cover crop system • CV, crownvetch living mulch system • NT, no-tillage system • OR, organic system • SADP, Sustainable Agriculture Demonstration Project
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Many of the improvements to soils as a result of no-tillage production are related to increases in soil organic C which in turn relates to improvements in soil aggregation, water-holding capacity, and nutrient cycling (Weil and Magdoff, 2004; Grandy et al., 2006). Soil organic C can also be increased by other strategies, including addition of winter annual cover crops into rotations, diversifying rotations with perennial crops, addition of manure-based amendments, and organic farming, which often employs all of the preceding strategies. For example, soil organic C and N were increased by both reducing tillage and using winter annual cover crops, leading the authors to suggest that the best management system would include no-tillage and a mixture of legume and nonlegume winter annual cover crops (Sainju et al., 2002). Rotations that included at least 3 yr of perennial forage crops had the highest soil quality scores with total organic C being identified as the most sensitive quality indicator (Karlen et al., 2006). Manure- and legume-based organic farming systems from nine long-term experiments across the USA were shown to increase soil organic C and total N compared with conventional systems (Marriott and Wander, 2006). Crop yields and/or soil organic C was increased by organic vs. conventional cropping systems in the East (Pimentel et al., 2005), Midwest (Delate and Cambardella, 2004), and West (Clark et al., 1998).
Most comparisons of soil improvements in organic vs. conventional cropping systems have been conducted under conventional tillage conditions. The dilemma for organic farmers is that the approaches for increasing soil organic C usually require tillage. Specifically, tillage is required for eliminating perennial legumes before rotation to annual crops, for incorporating manure to avoid N volatilization losses, or for preparing a seedbed and controlling weeds. Since an increase in tillage intensity and frequency has been shown to decrease soil C and N (Franzluebbers et al., 1999; Grandy et al., 2006), increases in organic matter by utilization of organic materials in organic farming may be offset by decreases in organic matter from tillage. Some authors have speculated that conventional no-tillage agriculture may provide superior soil improvement and potential environmental benefits compared with organic farming because of the tillage requirement of organic farming (Trewavas, 2004). The need for long-term research has been advocated to assess the relative merits of conventional no-tillage agriculture compared with organic farming (Macilwain, 2004). There is little literature reporting such long-term comparisons. One 6-yr study in Pennsylvania showed that some form of primary tillage was required for crop yields in organic systems to match those in conventional systems, but that a pure no-tillage organic system was not viable (Drinkwater et al., 2000).
A long-term experiment, the Sustainable Agriculture Demonstration Project (SADP), was initiated at Beltsville, MD, to compare selected no-tillage grain cropping systems and a reduced-tillage organic system on a sloping, droughty site typical of the mid-Atlantic piedmont. The standard for comparison was a no-tillage system typical of that used in this area. Two additional no-tillage systems, one including winter annual cover crops and another including a perennial crownvetch living mulch, were compared with this standard for their potential to improve soil organic matter, reduce external inputs, and enhance environmental protection on erodible soils. Finally, an organic cropping system that reduced tillage to the minimum necessary for incorporation of manure and for weed control was included in this comparison. Performance of these systems during the first 4 yr of the experiment, which included transition years for the organic system, was reported by Teasdale et al. (2000). A simulation of projected yields, economic returns, and environmental impacts was reported by Watkins et al. (2002). This paper reports results from a comparison of these systems over a 9-yr period as well as a 3-yr uniformity trial that followed.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Thirty-two plots that were 9.1 m wide and ranged from 120 to 170 m long were established on a contour across the slope of the field. Plots were separated by 1.8-m-wide grass strips. Rotations were initiated in fall of 1993 with establishment of wheat and cover crops. Four cropping systems were established, with each system initially following a 2-yr rotation with corn in the first year and wheat and/or soybean in the second year (Table 1). The intensive rotation with three crops produced in 2 yr (corn, wheat, and double-crop soybeans) is frequently practiced by growers in the mid-Atlantic region. On the basis of input from stakeholders in winter of 1998, changes were made to the crownvetch and organic rotations as described below.
|
Details of system operations were presented in Teasdale et al. (2000). In brief, lime, P, and K were applied according to annual soil testing of each plot. The P and K were applied in inorganic form to the NT, CC, and CV systems, but in organic amendments to the OR system. Levels of soil pH, P, and K were maintained at levels suitable for crop production throughout the study according to University of Maryland recommendations. Corn and full-season soybeans were planted in early May, wheat in late October, and double-crop soybean in early July. Corn was planted with a no-tillage planter in 76-cm rows at an average population of 66,700 seed ha–1. Organic soybeans were planted in 76-cm rows at an average population of 484,900 seed ha–1. Soybeans in the nonorganic systems were planted in 38-cm rows until 1999 and were drilled in 19-cm rows thereafter at an average population of 498,600 seed ha–1. Wheat was drilled in 19-cm rows at an average seeding rate of 130 kg ha–1 until 1999 and 223 kg ha–1 thereafter. Cultivars of each crop varied from year to year, but the same cultivar was planted in all systems in a given year.
System Description
The four cropping systems are described in Tables 1 and 2. All systems were developed with the expectation that (i) at least one grain crop would be harvested in every year, (ii) crops would be rotated, (iii) soil would be covered with vegetation or residue during as much of the rotation as possible, and (iv) tillage would be minimized to the extent possible within each system.
|
The CV system followed similar practices as the NT system except that crops were grown in a perennial crownvetch living mulch. This living mulch was expected to further reduce erosion and herbicide runoff compared with desiccated residue in a no-tillage field (Hall et al., 1984). Crownvetch was drilled after wheat harvest in 1994 and 1995 and allowed to establish during the remainder of those years rather than growing double-crop soybeans. Similar levels of inorganic N were applied to corn and wheat as the NT system since maximum corn yields were shown to require full N rates despite N contributions from the crownvetch (Duiker and Hartwig, 2004). Herbicides (Table 2) were applied before planting each crop with the intention to defoliate, but not kill, the crownvetch living mulch (N.L. Hartwig, 1995–2000, personal communications). Since excess competition with corn was experienced, a more aggressive program including a low dose of glyphosate at corn planting was initiated in 2000. Because no recovery time was available in the original corn–wheat/soybean rotation, the double-crop soybean crop was dropped from the rotation in 1998 to allow time for crownvetch regrowth.
The CC system used the winter annual cover crops hairy vetch before corn and rye before soybean. There was insufficient time to plant hairy vetch following a wheat/soybean double crop; thus, a full-season soybean crop was grown between May and September to permit hairy vetch to be planted in early October. Crops were planted into cover crop residue without tillage. Cover crop residue has been shown to suppress early-season weed emergence; consequently, a weed management program was followed that eliminated preemergence herbicides and applied postemergence herbicides based on species of emerged weeds (Teasdale and Rosecrance, 2003). In addition, hairy vetch has been shown to have a high N content and reduce the N requirement for corn (Decker et al., 1994); consequently, the preplant fertilizer N application to corn was eliminated and N was only banded with the planter and sidedressed as in the no-till system. Annual fertilizer N application to corn averaged 17 kg ha–1 with the planter plus 85 kg ha–1 sidedressed while N content of vetch aboveground biomass averaged 112 kg ha–1.
The OR system followed a 2-yr corn–wheat/soybean rotation through 1997. Beginning in 1998, this was expanded to a 3-yr corn–soybean–wheat rotation to provide a legume (soybean) before wheat, an earlier planting date for the crimson clover (Trifolium incarnatum L.) cover crop after wheat, and time after wheat to control late summer weeds and reduce weed seed inputs to the soil seedbank. This 3-yr rotation is typical of organic systems used in other long-term experiments (Pimentel et al., 2005; Marriott and Wander, 2006). Crimson clover was grown as a green manure crop for corn (aboveground biomass contained an average of 109 kg N ha–1). It was planted by overseeding into double-crop soybeans at leaf-drop before 1998 and thereafter by drilling into a chisel plow/disk-tilled seedbed in September following wheat harvest. Cow manure was applied to maintain soil nutrient levels based on University of Maryland recommendations and estimated N credits for the crimson clover cover crop. In 1994 to 1997, dairy solids were applied to corn and wheat at the average rate of 24,000 and 23,000 kg ha–1 (containing 0.51% N), respectively, and chisel plowed just before planting. Thereafter, composted dairy solids were applied to corn at an average rate of 9300 kg ha–1 (containing 1.3% N) by chisel plowing just before crimson clover seeding in September. Beginning in 1998, fresh dairy manure was applied to wheat at an average rate of 68,000 L ha–1 (containing 4.4 g N L–1) and chisel-plowed just before planting. Weeds were controlled in corn and soybeans until 1998 by chisel plow and disking before planting followed by rotary hoeing and sweep cultivation after planting. Beginning in 1999, corn and soybean were planted without tillage, crimson clover and rye were flail mowed, their residues were left on the soil surface to suppress weeds, and weeds emerging after crop establishment were cultivated with a high-residue cultivator.
Data Collection
Grain yield was determined by combine harvesting the middle eight rows of corn, five rows of 76-cm soybean, nine rows of 38-cm soybean, and 3.8 m of drilled soybean and wheat over the entire plot length and weighing harvested grain in a weigh wagon. Moisture content of grain was determined and yield adjusted to 15.5% for corn, 13.5% for wheat, and 13.0% for soybean. Remaining grain was harvested and wheat straw was baled from the entire plot area. Corn and soybean plant populations and wheat head density were determined at physiological maturity at sites where the four transects crossed each plot.
Percentage of soil area covered by vegetation was visually estimated at key times during the crop rotations. Vegetation cover by cover crops and winter vegetation was estimated in spring just before cover crop termination and corn and soybean planting operations. Vegetation cover by weeds was estimated at maturity (seed production) in corn and soybeans and after harvest of wheat. Estimates were conducted by dividing the length of each plot into four quadrants and visually estimating the total percentage of the middle 6 m of each quadrant that were covered by live vegetation other than the crop and then determining the species fraction of the total. In early spring, 7.5-cm-diam. soil cores were taken at 10 sites across the center of each plot for determining the weed seedbank. These samples were taken at depths of 0 to 10 and 10 to 20 cm in 1993 and 1994 and a portion of this soil archived in a –20°C freezer.
Soil was sampled with a 5-cm diam. corer at 16 sites within the center 6 m of each plot at depths of 0 to 7.5, 7.5 to 15, and 15 to 30 cm after summer crops were harvested in fall. Composite soil from each plot and depth was analyzed by the University of Maryland Soil Testing Lab for soil organic matter (loss on ignition), pH (1:1 water–soil slurry), and P, K, Ca, and Mg (Mehlich 3, Wolf and Beegle, 1995) for use in determining crop nutrient recommendations for the following season. Beginning in 1999, a portion of composite soil from each plot and depth was sent to Penn State Agricultural Analytical Services Lab to determine total C and total N by combustion using a Fisons (Milan, Italy) NA 1500 Elemental Analyzer (Bremmer, 1996; Nelson and Sommers, 1996). Archived soil seedbank samples were analyzed for total C and N by the Penn State Agricultural Analytical Services Lab in 2002.
Analysis of variance was performed with PROC MIXED (SAS v. 9.1, SAS Institute, Cary, NC) and mean separation using the PDIFF option of the LSMEANS statement at P < 0.05 when the F test was significant. The need for variance partitioning was determined by comparing the AIC, AICC, and BIC output values from PROC MIXED with and without partitioning. Partitioning was only required for the analysis of weed cover data.
Uniformity Trial
Operations for all cropping systems were terminated after crops were harvested in 2002. Plots were fallow over winter, and all vegetation including volunteer cover crops were destroyed with glyphosate in early spring to create uniform planting conditions across all plots. All plots were planted to corn in 2003 to 2005 with fertility and weed control managed according to the NT system (Tables 1 and 2). Two 9- by 9-m sections were established in each plot, one without fertilizer N but with herbicide and the other without herbicide but with N. Each minus-N section within a plot was established along one of the soil uniformity transects within that block and each minus-herbicide section was established along another transect within that block. These minus-N and minus-herbicide sections continued in the same location from 2003 to 2005. After corn emerged, subplots that were four rows wide and 6.1 m long were established where corn populations were uniform within the minus-N and minus-herbicide sections. Corresponding subplots were established in areas adjacent to each minus-N or minus-herbicide subplot where corn was grown with recommended N and herbicide practices. This provided within each plot a contrasting plus and minus N and plus and minus herbicide set of subplots, each established along a transect of uniform soil conditions within each block. This design allowed for relatively precise estimates of the 9-yr (1994 to 2002) cropping system historical effect on corn grown in 2003 to 2005.
In each subplot, corn population was determined after sidedressing N and grain yield was determined by hand harvesting the middle two rows at physiological maturity. Corn grain also was harvested from entire plots, except the minus-N and minus-herbicide sections by combine as described for the 1994 to 2002 period. Grain yield was adjusted for grain removed by hand-harvesting the plus-N and plus-herbicide subplots. A presidedress nitrate test (Bundy and Meisinger, 1994) was performed by taking soil samples to 30 cm before sidedressing from all plus- and minus-N subplots. Corn ear leaf N was determined at silking from all plus- and minus-N subplots. Weed cover ratings were conducted on all minus-herbicide subplots at weed maturity (all other subplots received herbicide and were kept weed-free by hand weeding if necessary). Analysis of variance was performed using PROC MIXED as described above.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
Weed cover in corn in the OR system was higher than that in systems with the best weed control in 6 of 8 yr and corn yield was lower in 4 of 6 yr (Table 3). Weed control and corn yield in the OR system were best in the first year of the experiment but declined as the experiment progressed. During 1994 to 1998, weed control by preplant chisel-plowing followed by rotary hoeing and cultivating maintained weed cover at <20% in 4 of 5 yr. Corn yields were similar in the OR and NT in two of those years, 1995 and 1998. In the other 2 yr, 1994 and 1996, lower yields in the OR system may have been the result of inadequate fertility to achieve the high yield potential resulting from good rainfall in those years. Spring chisel plowing was eliminated from 1999 to 2002 with reliance on cover crop residue and between row cultivation with a high-residue cultivator for weed management. Drinkwater et al. (2000) reported good weed control and yields equivalent to a conventional no-tillage system using a reduced-tillage organic system for corn production similar to this. However, this system resulted in poor weed control in 1999 to 2002 (Table 3) and a shift from a mix of smooth pigweed, common lambsquarters (Chenopodium album L.), and annual grasses in early years to domination by annual grasses, predominately giant foxtail (Setaria faberi Herrm.), in later years.
There were few differences between systems in harvested corn populations (data not shown). However, stand was reduced in OR in 3 of the first 4 yr (the 1994–1997 population reduction in OR vs. NT averaged 27%) and this may have contributed to yield reductions in those years. Stand losses in OR were the result of an uneven seedbed left after chisel plowing and disking the crimson clover cover crop and resulting damage by rotary hoeing and cultivating. Employment of the no-till planting system beginning in 1999 resolved these population problems.
Similar results were obtained in soybeans as corn but comparisons between systems were confounded by differences in planting dates among systems (data not presented). Full-season soybeans grown in CC averaged 3.38 Mg ha–1, whereas double-crop soybeans grown after wheat in NT averaged 2.02 Mg ha–1 across all years. Weed control was excellent in NT and CC soybeans in all years averaging <4% weed cover except for CC in 1994 and 1997, when morningglory (Ipomoea spp.) and annual grass weeds escaped to give 
20% year-end cover. Double-crop soybeans were grown in OR until 1997 and had similar yields to those in NT (P = 0.416) with weed cover that averaged 21% in those years. Full-season OR soybeans were grown without chisel plowing and rotary hoeing after 1998, resulting in poorer weed control that left an average of 39% year-end weed cover. Yield comparisons during these years were limited because of drought that resulted in no yields in 1999 and 2002 and rotation out of soybeans in 1998 and 2001.
Wheat yield was higher in the CV system than the NT system in 4 of 7 yr (Table 3). Since the crownvetch living mulch grows primarily in summer months, competition from this perennial during the winter wheat production season from fall to spring was minimal. But, N released from killed crownvetch before planting wheat probably stimulated tiller production and growth of wheat relative to the NT system. Wheat yield in the OR system was lower than that in the other systems in 3 of 7 yr. Lower yields may have been the result of insufficient N mineralization from fall manure application during cold season months. Also, volunteer rye in OR wheat in 2000 may have contributed to the lower yield in that year. Weed growth was minor in OR wheat during the growing season but weeds did increase in abundance after wheat senescence. The pattern of wheat yield response to cropping systems was usually similar to the pattern of wheat head density in these systems at maturity (data not presented).
At the end of the experiment, total combustible soil C and N made up a higher fraction of soil dry weight in the OR system than all other systems at all depths (Table 4). Soil C and N concentration was higher in the CC system than the NT and CV systems at the shallower depths. Soil organic matter determined by loss-on-ignition showed a similar pattern of response to cropping systems as that of total soil C (data not presented). At the beginning of the experiment, soil C and N was not determined. However, soils archived for seedbank analysis in 1993 and 1994 provided a means to determine initial total C and N levels. These samples were not taken at the same depths as soil nutrient samples, so a direct comparison of initial and ending values is not possible. However, analysis of these samples showed no initial difference among cropping systems in total soil C (P = 0.702 and P = 0.400 at 0–10 cm and 10–20 cm, respectively) or total soil N (P = 0.739 and P = 0.224 at 0–10 cm and 10–20 cm, respectively). This data provides evidence that differences between systems in soil C and N observed at the end of this experiment represented changes that occurred as a consequence of these systems during the course of this experiment.
|
|
In subplots with no herbicide application during 2003, weeds emerged earlier and more abundantly in the OR system, whereas the fewest weeds emerged and developed in the NT system (Table 6). As a result, yield reduction in the weedy subplots relative to the adjacent weed-free subplots was highest in the OR and lowest in the NT system in 2003. This result reflects the larger weed seedbank that probably had built up in the OR than the NT system over the preceding 9 yr. After 1 yr without herbicides, the seedbank apparently equilibrated in all subplots sufficiently so there were no differences in weed cover or yield reduction between systems in 2004 and 2005.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Winter annual cover crops can amplify many of the benefits of no-tillage cropping systems for protecting and improving soils. Cover crops offer several advantages for nutrient and water management (Decker et al., 1994; Clark et al., 1995). Previous research showed that the CC system had the highest grain production efficiency per unit N input and per soil water-use in this experiment (Teasdale et al., 2000). Higher vegetative residues (including more legumes) returned to soil in CC than in NT probably accounted for these higher N-use and water-use efficiencies (Teasdale et al., 2000) as well as the increase in soil C and N in CC vs. NT in the surface 15 cm of soil (Table 4). Seo et al. (2006) also showed that N from hairy vetch residue contributed about twice as much N to soil as did fertilizer N. However, overall soil conditions were not changed sufficiently in CC relative to NT to statistically affect corn yield in the uniformity trial following the systems trial (Table 5). Reliance on a postemergence herbicide program in CC also led to potentially minimal herbicide movement into ground and surface water (Watkins et al., 2002).
The CV system produced the highest average wheat yield and the highest corn yield in a year with exceptionally good moisture availability; however, competition between this perennial cover crop and corn led to excess competition for moisture when moisture was limiting. Previous research showed there was a reduction in water-use efficiency in CV compared with NT (Teasdale et al., 2000). Duiker and Hartwig (2004) also showed that crownvetch and other living mulch species reduced corn yields in dry years. Only when the herbicide program suppressed crownvetch to the point of stand depletion in the later years of this experiment was corn yield not affected. On the other hand, the CV system did provide long-term benefits to the soil as shown by increased subplot corn yield and greater N availability compared with NT during the uniformity years (Table 5). The CV system did not raise soil C and N concentrations during the 9 yr of this systems trail (Table 4), nor during a 10-yr period in a Pennsylvania trial (Duiker and Hartwig, 2004). Instead, increased N availability could be explained by development of the perennial crownvetch root system that opened new channels for subsequent corn roots to access N and water from deeper in the soil profile. Katsvairo et al. (2006) review the capacity of perennial grasses to increase the rooting depth of subsequent crops in rotation, particularly on shallow soils such as those at the SADP site.
The OR system resulted in lower average corn yield than the other systems and weed control declined as the experiment progressed (Table 3). Several factors contributed to poor weed control. First, chisel-tillage of crimson clover and manure led to an uneven seedbed and difficulty controlling early emerging weeds by rotary hoeing during the early years. Perron et al. (2001) also observed increased weed populations when chisel-plowing was combined with mechanical weed control because of surface residue interference with mechanical cultivation operations. Second, mowed cover crop residue used to suppress early weed emergence in later years was not sufficiently uniform to provide effective weed control (crimson clover averaged 74% soil coverage before mowing). Third, sweep cultivation controlled weed seedlings with a lower efficiency in no-tillage systems because of inability to fully disrupt weed root systems within the relatively larger soil aggregates than those found in tilled soil (Teasdale and Rosecrance, 2003). Fourth, the short grain crop rotation used in this OR system was probably unsuitable for maintaining sufficiently low weed seedbank populations for an effective organic weed control system. Longer rotations that include a hay crop are more effective at maintaining lower weed seedbanks and improving weed control in organic systems (Teasdale et al., 2004). Inclusion of a perennial hay crop in this organic rotation may have improved weed control and also provided additional yield-improving benefits as those observed from the perennial crownvetch cover crop described above.
Despite poor weed control, the OR system improved soil productivity significantly as measured by corn yields in the uniformity trial (Table 5). Teasdale et al. (2000) showed that, across the complete rotation, the OR system returned approximately twice the organic dry matter from manure (19 Mg ha–1) and vegetation (18 Mg ha–1) to soil compared with the NT, CC, and CV systems which returned 14, 20, and 16 Mg ha–1 of vegetation, respectively. This higher input of organic matter to soil in the OR system can account for the higher soil C and N concentration than all other systems after 9 yr (Table 4). These higher levels of soil C and N were achieved despite the use of tillage (chisel plow and disk) for incorporating manure and of cultivation (low-residue sweep cultivator) for weed control.
Conventional no-tillage production systems have been shown to provide important benefits to long-term soil sustainability, in part by eliminating soil and C losses caused by tillage operations. Our results suggest that systems that incorporate high amounts of organic inputs from manure and cover crops can improve soils more than conventional no-tillage systems despite reliance on a minimum level of tillage. Our results suggest that, if adequate weed control could be achieved in reduced-tillage organic systems, they would be capable of providing improved soil quality with yield-enhancing benefits compared with conventional no-tillage systems. On the other hand, our results also suggest that, if conventional no-tillage systems utilized additional organic inputs or rotational perennial crops, they could probably achieve yield-enhancing soil quality benefits as well.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. Senthilkumar, A. N. Kravchenko, and G. P. Robertson Topography Influences Management System Effects on Total Soil Carbon and Nitrogen Soil Sci. Soc. Am. J., October 21, 2009; 73(6): 2059 - 2067. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Senthilkumar, B. Basso, A. N. Kravchenko, and G. P. Robertson Contemporary Evidence of Soil Carbon Loss in the U.S. Corn Belt Soil Sci. Soc. Am. J., October 21, 2009; 73(6): 2078 - 2086. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Leifeld, R. Reiser, and H.-R. Oberholzer Consequences of Conventional versus Organic farming on Soil Carbon: Results from a 27-Year Field Experiment Agron. J., August 31, 2009; 101(5): 1204 - 1218. [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 | |||
