Agronomy Journal Grow Your Career With ASA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Web of Science (2)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Shapiro, C. A.
Right arrow Articles by Stroup, W. W.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Shapiro, C. A.
Right arrow Articles by Stroup, W. W.
Agricola
Right arrow Articles by Shapiro, C. A.
Right arrow Articles by Stroup, W. W.
Related Collections
Right arrow Other Crop Management
Right arrow Other Soil Management
Right arrow Tillage
Agronomy Journal 93:850-862 (2001)
© 2001 American Society of Agronomy

SOIL MANAGEMENT

Tillage and Management Alternatives for Returning Conservation Reserve Program Land to Crops

Charles A. Shapiro*,a, David L. Holshouserb, William L. Kranzc, David P. Sheltonc, John F. Witkowskic, Keith J. Jarvic, Gerald W. Echtenkampa, Lisa A. Lunza, Robert D. Frerichsa, Ray L. Brentlingera, Mari A. Lubberstedta, Melinda McVey McCluskeyd and Walter W. Stroupe

a Haskell Agric. Lab., Univ. of Nebraska, 57905 866 Rd., Concord, NE 68728-2828
b Virginia Tech, Tidewater Agric. Res. and Ext. Cent., 6321 Holland Rd., Suffolk, VA 23437
c Northeast Res. and Ext. Cent., Univ. of Nebraska, 601 East Benjamin Avenue, Suite 104, Norfolk, NE 68701-0812
d Northeast Community College, East Benjamin Ave., Norfolk, NE 68701
e Dep. of Biometry, Univ. of Nebraska, Lincoln, NE 68583

* Corresponding author (cshapiro{at}unl.edu)

Received for publication January 5, 2000.

   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Accumulated vegetative residue was a concern when Conservation Reserve Program (CRP) land returned to grain crop production. This study was conducted to determine the effect of residue management, tillage, and crop choice on grain yield in the first year of cropping on CRP land that was predominately smooth brome (Bromis inermis Leyss). Three residue management practices (undisturbed, shred, and remove), three tillage systems [moldboard plow, disk, and no till], and three crops {corn (Zea mays L.), soybean [Glycine max (L.) Merr.], and grain sorghum [Sorghum bicolor (L.) Moench]} were used in a factorial arrangement of a 3-yr field experiment conducted in Nebraska on fine-silty, mixed, mesic Udic Haplustoll; fine-silty, mixed (calcareous), mesic Typic Ustorthent; and fine-silty, mixed, mesic Cumolic Halustoll soils. Residue management was not significant for corn (P > F = 0.16), sorghum (P > F = 0.113), and soybean (P > F = 0.491) although there were significant residue x tillage interactions. Tillage system was not significant (P > F = 0.125) for soybean yields, but plowing significantly (P > F = 0.0001) increased both corn and sorghum yields. Mean corn yields were 13% less for the no-till system than for the moldboard plow system. However, no-till corn yield differences were not significant (P > F = 0.255) when plant population (a possible measure of planter performance) and percent green rating (a measure of weed control) were included as covariates. Our recommendation for the first year of grain crop production on smooth brome CRP land is to shred the residue and plant soybean in a no-till system.

Abbreviations: CRP, Conservation Reserve Program • DK, disk treatment (three disk treatments used before planting) • NT, no-till treatment (no mechanical seedbed preparation used before planting) • PC, percent • PL, plow treatment (moldboard plow followed by three disk treatments) • RMV, remove treatment (smooth brome mowed, baled, and removed before tillage) • SHD, shred treatment (smooth brome shredded, but remaining on the ground, before seedbed preparation) • UND, undisturbed treatment (no mechanical disturbance)


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
THE CONSERVATION RESERVE PROGRAM (CRP) was authorized as part of the 1985 Food Security Act (USDA, 1992) to remove environmentally sensitive and highly erodible land from crop production and place it in perennial vegetative cover. Between 1986 and 1992, the USDA developed 10- and 15-yr contracts for 14 million ha of highly erodible land (USDA, 1992). Specific vegetative cover varied by region of the country and landowner choice. Although many grass cover mixtures were permitted in northeast Nebraska, the two main grass systems were cool-season smooth brome or warm-season switchgrass (Panicum virgatum Leyss).

Management of CRP ground during the contract period was often minimal. Except under emergency circumstances, contract holders were not allowed to harvest or graze the forage produced. Therefore, most CRP land received no fertilizer, minimal weed control, and little vegetation removal. A Minnesota survey of 151 CRP fields found little change in soil pH, P, or K at the end of CRP contracts compared with the beginning (Jewett et al., 1996a). A 90% infestation of weeds was also found in existing CRP fields (Jewett et al., 1996b). Thus, returning CRP land to row crop production differs from bringing hay or pastureland into production. Land used for hay or pasture may be easier to crop because less plant residue is accumulated, and periodic cutting may reduce weed infestations.

One of the stated purposes of the CRP program was to improve soil physical and chemical properties. Ten years of grass production without residue removal would reduce erosion and create improved conditions for subsequent cropping. Soil organic C levels in the top 0.4 m at five sites adjacent to CRP land in Texas, Kansas, and Nebraska indicated an average increase in soil organic C of 4.0 Mg ha-1 after 5 yr (Gebhart et al., 1994). Krall and Schuman (1996) reviewed the literature on integrated dryland crop and livestock systems and concluded that sustainable systems need to be developed on a site-specific basis to provide alternatives to continuous cropping. Maintaining a hay or grazing system on former CRP land would minimize erosion and runoff (Gilley et al., 1996). Conversely, the tillage and planting systems used to initiate row crop production on CRP acres will generally increase erosion and runoff.

Early national surveys estimated that 68% of CRP land would go back into crop production (Osborn et al., 1994). Northeast Nebraska producers were projected to return the greatest amounts of CRP land to cropland compared with all of the other Agricultural Statistics Districts in the state (Clark et al., 1994). Thus, research was needed to determine the consequences of returning CRP land to crop production. Laryea and Unger (1995) reported that, under moisture-limiting conditions, no tillage on grassland converted to crops increased yields and maintained soil organic C compared with sweep tillage or moldboard plow. Rainfall simulation research showed that undisturbed CRP land tended to have greater initial water runoff rates, but after 64 mm of rainfall, undisturbed CRP land had less water runoff and soil erosion than some tilled treatments (Gilley et al., 1997; Gilley and Doran, 1997). The fine root mass from the CRP grass initially kept tilled soils from being immediately susceptible to erosion. However, when measured following 9 mo of fallow after tillage, erosion increased and the tilled CRP land had the same erosion rate as continuously cropped soils (Gilley and Doran, 1998). The emerging concern was how to economically bring CRP land into production without losing the soil quality benefits that had been achieved. An economic analysis following tall fescue [Festuca arundinacea (L.) Schreb] showed that continuous no-till systems provided the highest net income and continuously plowed systems had the lowest net income over a 6-yr period (Phillips et al., 1997). Nitrate leaching was decreased during CRP (Randall et al., 1997). Tile drainage was 5.3 times greater for row crops than for CRP land in years of excess precipitation, and the flow-weighted average NO3–N concentration was 32 mg L-1 for continuous corn compared with 2 mg L-1 for CRP land (Randall et al., 1997).

Plowing CRP land could be an effective method for killing grasses and allowing crops to be planted the first year. Because of the large quantity of accumulated plant residue, it was unknown what effect the undisturbed residue would have on subsequent crops. Large quantities of high C residue incorporated in the soil might immobilize N. Disking before and after moldboard plowing would prepare a seedbed that was suitable for planting. This intensive tillage might reduce soil quality and leave bare soil surfaces open to erosion. Reduced tillage maintains more cover and soil quality but might not allow profitable production if herbicides fail or a uniform plant stand is not established. Even under reduced tillage, planting might be easier if residue was reduced or removed. Various residue reduction techniques, such as burning, shredding, or haying, could be used to allow easier tillage and more effective weed control.

The specific crop planted in the first growing season could interact with tillage and residue management because crops have varying fertility requirements, drought resistance, and ease of establishing uniform stands. In addition, more herbicide choices may allow one crop species to be planted in a cleaner seedbed.

Because of the interrelated concerns of residue management, tillage, and crop choice, the objective of this study was to determine the effects of each of these factors on grain yield in the first year of cropping after smooth brome on CRP land.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
The experiment was conducted on a 65 ha farmer-owned CRP field, 2.4 km east of Concord, NE and 40 km west of Sioux City, IA. Soils at the rainfed experimental site consisted predominantly of Moody and Nora silty clay loam (Fine-silty, mixed, mesic Udic Haplustoll), Crofton silt loam [Fine-silty, mixed (calcareous), mesic Typic Ustorthent], and Alcester silt loam (Fine-silty, mixed, mesic Cumulic Haplustoll) (USDA, 1978).

The experimental site was in alfalfa (Medicago sativa L.) before being placed in CRP in 1986, at which time it was planted to smooth brome. No field operations were conducted during the time that the field was in CRP. Smooth brome, with scattered patches of alfalfa, was the predominate vegetation at the time of experiment initiation in October 1994. At that time, three areas, each approximately 12 ha, were designated as the experimental units of the field that would be returned to grain crop production in 1995, 1996, and 1997, respectively (Fig. 1).



View larger version (51K):
[in this window]
[in a new window]
  		Fig. 1. Schematic of field layout by residue blocks for each experiment.

 
A factorial arrangement of three residue management practices, three tillage systems, and four crop rotations was initiated on CRP land in each of the 3 yr. A split-split plot experimental design with four replications was used each year, with the three residue management treatments as the main plots, the three tillage treatments as the subplots, and the four 3-yr crop rotations (continuous corn, corn–soybean–corn, soybean–corn–soybean, and grain sorghum–soybean–corn) as the sub-subplots (12.2 by 12.2 m). The three residue management treatments were: (i) undisturbed (UND), nothing more than herbicides used to manage the smooth brome stand before tillage, i.e., no mechanical disturbance used; (ii) shred (SHD), smooth brome shredded, but remaining on the ground, before seedbed preparation; and (iii) remove (RMV), smooth brome mowed, baled, and removed before tillage. The three tillage treatments were: (i) plow (PL), moldboard plow followed by three disk treatments to prepare the seedbed; (ii) disk (DK), three disk treatments used before planting; and (iii) NT, no mechanical seedbed preparation used before planting. Although the crop treatment was a 3-yr rotation, only the results of the first year of each rotation are reported in this paper. A new randomization was used each year on a new area being removed from CRP. The three residue management treatments are for the first year out of CRP only.

Each crop was analyzed separately with a combined analysis over years. Analyses were computed using SAS PROC MIXED (SAS Inst., 1997). Where residue x treatment interactions were statistically significant (P < 0.05), simple effect differences were evaluated among residue levels for each tillage treatment and among tillage treatments for each residue level. Simple effect comparisons were obtained using the LSMEANS SLICE option in PROC MIXED. All analyses used Satterthwaite's procedure to determine appropriate error degrees of freedom (Littell et al., 1996). Treatments were compared based on differences among least square means and their associated standard errors. When missing data affected the standard errors of differences among treatments, the largest standard error in the set of comparisons is reported.

Biomass samples were collected from two undisturbed 0.84-m2 areas in each residue management treatment block before the start of each year's experiment. All aboveground vegetation was removed and partitioned as standing residue and accumulated mulch at the soil surface. Collected material was dried to constant weight, ground to pass through a 2-mm screen on a Wiley mill, and analyzed for total C and N (Schepers, 1989).

Soil samples were collected each fall before initial treatments were applied from the UND experimental units in each replication. Average soil chemical analysis indicated that soil P for most treatment blocks was above established critical levels for corn and sorghum of 15 mg kg-1 Bray no. 1 extractable P (Table 1). A critical level of 0.8 mg kg-1 diethylenetriamine pentaacetic acid (DTPA) is used for Zn. Soil K levels were also above the critical level of 125 mg kg-1 (KCl extractable) for corn (Hergert et al., 1995) and sorghum (Sander and Frank, 1980). Critical levels of 10 mg P kg-1 (Penas and Wiese, 1987) and 125 mg K kg-1 are used for soybean.


View this table:
[in this window]
[in a new window]
  		Table 1. Initial soil chemical measurements and aboveground biomass of Conservation Reserve Program (CRP) land in smooth brome returning to crop production.

 
In order to remove potential P and Zn deficiencies in areas with undetected low P and Zn levels, a liquid mixture of ammonium polyphosphate (10–15–0) and ammoniated Zn (20%) was applied to all corn and sorghum treatments. A band 10 mm from the seed furrow was applied at planting at rates of 9, 10, and 0.56 kg ha-1 N, P, and Zn, respectively, in 1995 and 1997. In 1996, a similar solution was used, except no Zn was included due to an operator mistake.

In all years, the NT and DK tillage systems were sprayed in the fall with glyphosate [N-(phosphonomethyl)glycine] to kill the existing grass vegetation. The PL treatment killed the perennial weeds and grass. Dicamba (3,6-dichloro-2-methoxybenzoic acid) and/or 2,4-D [(2,4-dichlorophenoxy)acetic acid] were also applied in the fall to help control perennial broadleaf weeds. In 1994, dicamba was tank-mixed with the glyphosate in the NT and DK plots. Due to heavier hemp dogbane (Apocynam cannabinum L.) and common milkweed (Asclepias syriaca L.) infestations in 1996 and 1997 crop years, dicamba and/or 2,4-D were applied as a separate application to all plots. Better weed control could be obtained by applying dicamba and/or 2,4-D before glyphosate when the weeds were actively growing and before cold temperatures slowed growth (Martin et al., 1998). Details of herbicides and field practices are given in Tables 2, 3, and 4.


View this table:
[in this window]
[in a new window]
  		Table 2. Summary of residue and tillage management associated with returning Conservation Reserve Program (CRP) land to crop production.

 

View this table:
[in this window]
[in a new window]
  		Table 3. Summary of cultural practices associated with returning Conservation Reserve Program (CRP) land to crop production.

 

View this table:
[in this window]
[in a new window]
  		Table 4. Summary of preemergence and postemergence weed management operations for returning Conservation Reserve Program (CRP) land to crop production.

 
Urea ammonium nitrate (28–0–0) solution was applied preplant to the corn and sorghum treatments with a Blue Jet 5-knife (Blu-Jet Products, Thurston, NE) applicator equipped with coulters. Knives were operated at a depth of 150 mm on 0.76-m spacings. Sorghum received 135, 112, and 112 kg N ha-1 in 1995, 1996, and 1997, respectively, and corn received 135, 168, and 168 kg N ha-1 in 1995, 1996, and 1997, respectively. Nitrogen for the corn was increased in 1996 and 1997 based on a N rate study conducted concurrently. Soybean was inoculated each year, but no fertilizer was applied.

Corn and sorghum were planted in 0.76-m-wide rows with a four-row John Deere Max-Emerge Model 7000 planter (Deere & Co., Moline, IL) in 1995 and 1996 and an eight-row John Deere Max-Emerge Model 7100 in 1997. Soybean was planted with a 4.6-m John Deere Model 750 drill at 0.19-m row spacing in all 3 yr.

Residue cover was measured at a randomly selected location in each of the 144 individual treatment plots approximately 2 wk after planting in 1997 using the line-transect method as described by Shelton et al. (1993). Percent (PC) residue cover was log10 transformed before statistical analysis was conducted.

Emergence counts were taken in 1996 and 1997, approximately 2 wk after initial emergence by counting at least 3.0 m of row for corn and sorghum and 0.84 m2 for soybean. In 1995, one population count was taken at the end of June, approximately 3 wk after planting, and was used to calculate emergence and final population. In 1996 and 1997, harvest population was determined by counting one row within the harvest area for each crop and each combine pass. In 1997, 0.84 m2 was counted per combine pass for soybean.

Visual ratings of the PC green cover in each experimental unit were made in the spring of each year to determine the effectiveness of the glyphosate treatments in killing smooth brome. The PC green rating reflected the PC ground area showing green smooth-brome regrowth. Because the PC green rating in 1995 was made before plowing, the PC green for the PL treatment was assumed to be zero for that year. Percent green was log10-transformed before statistical analyses were conducted. Log-transformed PC green was used as a covariate to analyze grain yield.

Monthly precipitation was measured from daily observations at a Department of Commerce weather station located 0.8 km from the experimental site (Fig. 2). Average temperature data were obtained from the Nebraska Automated Weather Station network located 1.0 km from the experimental area (Fig. 2).



View larger version (53K):
[in this window]
[in a new window]
  		Fig. 2. Average monthly rainfall from 1995 through 1997 and 25-yr average and average monthly temperature from 1994 through 1997 and 16-yr average, Haskell Agric. Lab., Concord, NE.

 
Soil temperature at a 0.1 m depth was monitored using Weksler (Weksler Instrument, Freeport, NY) thermometers in 1995 and Taylor (Taylor Instruments, Chicago, IL) thermometers in 1996 and 1997. Thermometers were placed in the RMV (1995) and SHD (1996 and 1997) residue treatments of the soybean and corn treatments. Thermometers were placed in the NT and PL treatments in all years and in the DK treatments in 1996 and 1997. Temperatures were recorded at least every 3 d from crop emergence to complete canopy and then weekly until the end of July (corn tassel emergence). Crop growth stage was recorded from selected tillage treatments on a biweekly basis (Ritchie et al., 1996; Ritchie et al., 1982; Vanderlip, 1979).

Soil moisture was monitored using a Campbell Pacific (CPN Int., Martinez, CA) Model 503DR neutron moisture probe. Aluminum access tubes were installed 1.8 m deep in three replications of the corn, sorghum, and soybean rows after plant emergence in 1996 and 1997. In 1995, access tubes were only installed in corn and soybean. Access tubes were placed in all tillage plots of the SHD and RMV residue treatments. Readings were made in 0.3-m increments every 14 d until after physiological maturity. Volumetric soil moisture was calculated from neutron scattering counts by using a locally established calibration curve.

Harvest was conducted using plot combines designed for each crop [Massey (Agco Corp., Duluth, GA) Model 10 for soybean and sorghum; Massey-Ferguson Model 300 (Kinkaid Equipment Manufacturing, Haven, KS) for corn]. Crop yield was measured from 6 rows (4.6 m) for corn and sorghum and 3.0 m from soybean. Grain moisture was measured at harvest and adjusted to 155, 140, and 130 g kg-1 moisture for corn, sorghum, and soybean, respectively.


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Smooth Brome Biomass
For all 3 yr, standing residue at time of treatment initiation averaged 3.13 Mg ha-1 in the UND treatments, and the surface mulch averaged 10.6 Mg ha-1 (Table 1). The RMV treatment harvested approximately 2.24 Mg ha-1 of the standing material. The surface mulch was partially decomposed as indicated by the C/N ratios in the 20:1 to 30:1 range compared with the standing residue that had C/N ratios of 35:1 to 50:1. The N content of the surface mulch was in the typical range reported for corn, sorghum, and soybean residue of 9.5, 13.5, and 13.5 g kg-1, respectively (Parnes, 1986).

The quantity of residue and how it is managed affects a number of factors, including the processes of N immobilization and mineralization. Removal of 2.24 Mg ha-1 of high C standing residue may affect immobilization and reduce total soil C and N. Surface mulch may also influence soil temperature, soil moisture, and herbicide efficacy. Differences in residue quality and particle size may exist between smooth brome and row crop residue, affecting these processes even with similar C/N ratios. The lower C/N ratios in the accumulated material indicates that it has partially decomposed and may be resistant to further decomposition compared with fresh corn or soybean residue (Prasad and Power, 1997).

Crop Development
The cropping years of 1995 and 1996 were wetter than normal over the whole season, but rainfall was not evenly distributed (Fig. 2). Average mean temperatures were cooler than normal for April and May in all years (Fig. 2). All crops developed faster in the PL treatments vs. the NT treatments (data not shown). In 1995, plants in the PL treatments were 0.3 to 1 leaf stage ahead of NT corn until mid-August. Soybean showed a similar trend, a tendency that widened as the season progressed. Sorghum plant stage was not monitored in 1995. In 1996, a similar relationship existed between PL and NT corn: The NT treatments developed more slowly than the PL treatments. However, the difference narrowed by early-August. Soybean plant development differences between NT and PL were evident early, but by early July, growth stage was the same. Sorghum plants in the PL treatment were ahead of NT by one leaf stage until mid- July. After mid-July, the differences were only a fraction of a leaf stage. In 1997, the same trends were evident in all crops: Any growth differences narrowed by August. For sorghum, the plants in the PL treatments were consistently about one leaf stage ahead of those in NT through August.

Percent Residue Cover after Planting
Residue management and tillage treatment significantly interacted in their influence on PC cover remaining after planting (Table 5). Although a disk tillage system is often thought of as conservation tillage, when the residue was removed, only about 15% cover remained, which did not meet the 30% cover criterion to be classified as conservation tillage (Conserv. Tillage Inf. Cent., 1986). Substantial cover remained with NT and DK with SHD or UND residue management treatments.


View this table:
[in this window]
[in a new window]
  		Table 5. Percent residue cover after planting on land coming out of Conservation Reserve Program (CRP).

 
Emergence and Plant Population
The proportion of emerged plants to seeding rate was generally lower than what is considered normal for all crops. Because planting into CRP land might affect emergence percentages, planted seeds per hectare were slightly greater than normal to compensate for any potential losses. For corn, PC emergence and harvest population were less for NT than PL in 1996 and less than DK and PL in 1997 (Fig. 3a and 3b). There were no PC emergence or harvest population differences in 1995. Similarly, sorghum PC emergence and harvest populations were lower under NT and DK compared with PL in 1996 and 1997 (Fig. 3c and 3d). Sorghum PC emergence for both NT and PL were greater than DK in 1995. Corn and sorghum harvest populations were on the low side of acceptable populations for rainfed production in the area. For soybean in 1995, NT had greater PC emergence than DK or PL. In 1996, NT PC emergence was intermediate and, in 1997, it similar to DK but less than PL (Fig. 3e). The NT harvest populations were greater in 1995 than those of DK or PL and similar to those of PL in 1996 and 1997 (Fig. 3f). Compared with recommended soybean populations in conventional agriculture of 309 to 432000 plant ha-1, harvest populations for soybean were 218000 plants ha-1 in 1997.



View larger version (52K):
[in this window]
[in a new window]
  		Fig. 3. Effect of residue and tillage on plant population and emergence of corn, sorghum, and soybean for land coming out of Conservation Reserve Program (CRP). DK, disk treatment; NT, no-till treatment; and PL, plow treatment. Error bars indicate standard errors.

 
Soil Temperature and Moisture
Average soil temperatures were consistently greater in the PL treatments compared with NT treatments for both corn (data not shown) and soybean (Fig. 4). Before crop canopy development, incoming solar radiation determined the magnitude of the soil temperature differences. In 1995, although no significant differences existed, mean soil temperatures for the PL treatment were typically 3 to 5°C greater than those for the NT treatment (Fig. 4a). In 1996 and 1997, soil temperatures were 1 to 3.5°C greater in the PL treatments compared with the NT treatments (P < 0.05; Fig. 4b,c). Temperature differences of this magnitude can affect germination and early crop development (Schneider and Gupta, 1985).



View larger version (19K):
[in this window]
[in a new window]
  		Fig. 4. Soil water content response to no-till (NT) and plow (PL) tillage treatments for soybean following Conservation Reserve Program (CRP). Error bars indicate standard errors.

 
Soil water content recorded for a 1.8-m soil profile showed no significant differences within years by crop or tillage. However, the PL treatment typically had lower water contents than NT (Fig. 4). Soil water content in 1997 steadily decreased due to below-normal May through August rainfall (Fig. 2).

Corn Yield Analysis
Residue management did not significantly affect corn yields over the 3 yr (Table 6). However, tillage system did affect yields in all years, with lower yields produced with NT. While NT had the lowest yields, the relationship among the three tillage treatments varied among years. In the combined over-years analysis for corn yield, there was a significant year x tillage interaction. In addition, there is evidence of tillage-dependent residue effects, or alternatively, residue-dependent tillage effects, as indicated by the fact that several of the simple effect comparisons within the tillage x residue interaction were significant (Table 6).


View this table:
[in this window]
[in a new window]
  		Table 6. Mixed model results from over-years analysis for the effect of residue and tillage on yield for land coming out of Conservation Reserve Program (CRP).

 
The least square mean year x tillage interaction yields are shown in Table 7. Corn yield averaged 14% greater with PL compared with NT in each of the 3 yr. The interaction among years was due to the changing relationship between the DK treatment and the NT and PL treatments. In 1995 and 1997, the DK and PL yields were similar, but in 1996, DK yield was less than that of the PL treatment.


View this table:
[in this window]
[in a new window]
  		Table 7. Effect of tillage and covariate analyses on corn grain yield on land coming out of Conservation Reserve Program (CRP).

 
Table 8 presents the pairwise yield differences between tillage treatment means for each residue treatment and the pairwise yield differences between residue treatment means for each tillage treatment. As indicated in Table 6, the significant simple effects in the no-covariate column for corn are tillage given UND, tillage given SHD, and residue given NT. Table 8 indicates which pairs of treatments within simple effect comparisons are significantly different and the magnitude of their estimated difference. The main conclusion is that residue management method had the greatest effect with the NT tillage treatment. If corn is to be no-till planted, residue should be removed. For the other tillage treatments, residue had no statistically significant effect.


View this table:
[in this window]
[in a new window]
  		Table 8. LSMEAN yield differences without the covariate analysis for the effect of residue and tillage treatments on corn, grain sorghum, and soybean yield, 1995–1997.

 
Concluding that NT was not appropriate for CRP conditions should not be done too hastily. The differences in PC residue cover after planting in this experiment (Table 5) indicate a much greater soil erosion potential for PL than for NT. For example, on northeast Nebraska soils and slopes comparable to those in this experiment, Dickey et al. (1984) reported predicted soil losses of 25.0 and 0.01 Mg ha-1 following 51 mm of water application for corn residue covers of 2 and 95%, respectively. Laryea and Unger (1995) showed that, under moisture-limiting conditions, no till increased yields and maintained soil organic C compared with sweep tillage or moldboard plowing when returning grasslands to crop production. Further, Gilley et al. (1996), Gilley and Doran (1997), Gilley et al. (1997), and Gilley and Doran (1998) documented increased soil losses from plowed CRP fields.

The yield loss associated with NT corn on CRP land may have been due to management decisions that could be changed. Plant population is one management variable that should be considered and could partially account for reduced NT yield. Emergence and harvest population counts (Fig. 5) indicate the problems we experienced achieving target corn plant populations in all 3 yr. In both 1996 and 1997, plant populations for NT were lower than those for the PL treatments. The reduced populations for NT may have been partially due to a lack of planter adjustment, specifically for the difficult conditions encountered.



View larger version (22K):
[in this window]
[in a new window]
  		Fig. 5. Soil temperature response to no-till (NT) and plow (PL) tillage treatments for soybean following Conservation Reserve Program (CRP). Error bars indicate standard errors.

 
Although reduced plant population may have affected yield, covariate analyses using population as the covariate indicated that this was not necessarily the case in all years (data not shown). Least square means adjusted for population differences showed no differences in corn grain yield due to tillage system in 1997, but even with adjustments for harvest population, there were still differences between the NT and PL systems in 1995 and 1996. The 3-yr average yield difference between NT and PL narrowed from 897 to 558 kg ha-1 when harvest population was included as a covariate. Therefore, some but not all of the yield reduction in NT can be explained by population differences.

Visual observations early in the season showed stunted plants lacking a dark green color in the NT treatments. Temperature data indicated cooler soils during the spring in the NT treatments, which may have adversely impacted plant development and yield potential. Soil moisture differences were not consistently different among tillage treatments.

The implications of our data are contrary to some recommendations that encourage the use of no till when returning CRP land to crop production (Johnson and Quarles, 1998). However, Phillips et al. (1997) reported yields significantly less for no till than for moldboard or chisel plow in the first year following CRP. After the fifth year of cropping, corn yields were greater for no till than for moldboard plow. Our NT treatments used slot planting without any zone tillage or residue removal. Our UND and SHD residue treatments were also extreme in that all of the residue was left on the surface, most of it as a mulch. Under NT with RMV, corn yields were 659 kg ha-1 greater than with UND (Table 8). The DK treatments were intermediate in the degree of residue disturbance between PL and NT, with yields generally reflecting these intermediate conditions.

Another reason that corn grain yields could have been reduced was competition from smooth brome escapes, which a more effective herbicide program may have prevented. Visual observation indicated that in-season weeds were mostly under control. However, smooth brome escapes were a problem in the NT treatments. Average PC green ratings were 22, 12, and 5 for NT, DK, and PL, respectively, indicating increased competition by smooth brome in NT. Percent green rating was also used as a covariate in the corn yield analyses, which showed significant year and tillage effects (data not shown) similar to the population covariate. With the covariate analysis of PC green rating, the adjusted differences in least square means between the PL and NT treatments were narrowed to 440 kg ha-1. The year x tillage interaction was significant because there was no difference in yield in 1997, but in 1995 and 1996, there was a difference of 377 and 1005 kg ha-1, respectively.

The PC green rating was used as a covariate with corn harvest population to analyze corn grain yield. By using both covariates, the significant tillage effect disappears (Table 6), but there was still a year x tillage effect. In 1995 and 1996, the difference between PL and NT decreased from 634 to 389 kg ha-1 and from 1468 to 878 kg ha-1, respectively, compared with the no-covariate analysis (Table 7). In 1997, there were no differences due to tillage. Also, there was no significant interaction between the two covariates in any year or overall.

Utilizing the covariate analysis reveals two main implications when returning CRP land to corn production. First, if similar stands can be achieved by planter adjustments or residue management in reduced tillage systems, yields will improve to levels near those of conventional tillage systems. Second, if the vegetation can be more completely killed, yields will also improve. Most importantly, these data reveal that yields similar to those of conventional tillage can be obtained with no- or reduced tillage systems if precautions are taken to ensure adequate stands and vegetation control.

Sorghum Yield Analysis
Sorghum and corn yields followed similar trends with reduced yields in NT. Over the 3 yr, NT and DK treatments yielded 83.4 and 93.4% of PL treatments, respectively. Unlike corn, sorghum had no year x tillage interaction. Sorghum harvest population was not a significant covariate for sorghum yield. Within each residue treatment, differences in yields were found among tillage treatments. Table 8 lists all six slice effects with the yield differences within each residue or tillage main effect. Within the NT treatments, the UND treatment reduced yields 1249 kg ha-1 compared with RMV and 950 kg ha-1 compared with the SHD treatment. Within both the DK and PL systems, residue treatment made no difference although the UND treatment tended to have lower yields. When there was tillage, the residue treatment made very little difference.

Covariate analysis with harvest population was not significant in contrast to the covariate analysis with PC green. Including both harvest population and PC green increased the probability of the F statistic for the tillage effect from 0.0001 to 0.0617. The difference in least square means between PL and NT was reduced from 1035 to 348 kg ha-1 (Table 9). The slice effect of varied tillage within both RMV and SHD residue managements was not significant with the PC green rating in the model. Covariate analysis indicates that differences in population were not as important for sorghum yield as they were for corn. Smooth brome competition, as indicated by the PC green rating, did, however, influence yields. Improved smooth brome control in NT would reduce the yield advantage of plowing seen in this study.


View this table:
[in this window]
[in a new window]
  		Table 9. Effect of tillage and covariate analyses on sorghum grain yield on land coming out of Conservation Reserve Program (CRP).

 
Soybean Yield Analysis
Average yields were 2450, 3513, and 2177 kg ha-1 for 1995, 1996, and 1997, respectively (Table 6). There were no year x treatment interactions. Neither the main effects of residue management nor tillage affected yields. Soybean did not show the same yield reduction in NT as was found in corn and sorghum. The average yields were 2663, 2691, and 2786 kg ha-1 for NT, DK, and PL, respectively. Covariate analysis with soybean harvest population and PC green rating, either together or separately, was not significant.

The residue management x tillage interaction was not significant, but the simple effect of the UND (with varied tillage) was significant (P > F = 0.0323; Table 6), with average yields of 2543, 2670, and 2834 kg ha-1 for UND-NT, UND-DK, and UND-PL, respectively. The DIFF option in PROC MIXED identified the significant comparison within the UND treatments. The difference was between NT and PL, amounting to 291 kg ha-1 (Table 8). This result for soybean was the same as for corn: Yields were reduced for the UND-NT treatment compared with the other treatments.

An examination of the soybean yields by year indicates that the simple effects associated with UND-NT treatment combination were due to significant differences in 1997 (data not shown). In 1997, the UND-NT soybean yields (1681 kg ha-1) were reduced below the UND-DK (1989 kg ha-1) and UND-PL (2284 kg ha-1) yields. In addition, the 1997 UND-NT soybean yields (1681 kg ha-1) were reduced below the SHD-NT (2519 kg ha-1) and RMV-NT yields (2361 kg ha-1). Over the 3 yr, the SHD-NT yields (2816 kg ha-1) were no different than the SHD-DK (2807 kg ha-1) or SHD-PL (2744 kg ha-1) yields. Shredding or removing residue before killing the grass would avoid the 1 in 3 yr where the UND-NT soybean yields were below the others.


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
Crop growth and yield, especially for corn and sorghum, was mostly affected by tillage treatment. Crop yields were not significantly influenced by residue management. Removal or shredding of the residue made field operations easier and increased herbicide efficacy, but yield differences due to residue treatment were only documented within NT systems.

The claim could be made that the population differences in NT confounded yields and biased results. However, the evidence only partially supports this view. Emergence counts were actually greater for NT corn and sorghum in 1995, and DK and NT had similar harvest populations in 1996. Yields were reduced for NT compared with DK even at similar populations. Covariate analysis did increase the adjusted NT yields relative to PL yields, but NT yields were still significantly less.

Returning CRP land to crop production is a difficult task, requiring planning and timely field operations. Effective cool-season grass control is facilitated by herbicide use the previous fall when adequate moisture is available for fall growth. Cutting and shredding standing grass in late summer and allowing regrowth before spraying herbicide will improve grass kill. Additionally, smooth brome control might be improved through increasing glyphosate application rates to obtain a better perennial grass kill. Using PC green ratings as a covariate with population reduced the NT and PL difference in corn yield to 364 kg ha-1 overall. Population improvements may be made through planter adjustments. However, even with the combined covariates, increased yields resulted with plowing in 2 out of 3 yr.

The yield differences were not significantly different among tillage systems for soybean. For the SHD or RMV conditions, NT soybean yields were equal to other tillage systems. There are a number of reasons why the tillage effect is less important in soybean: A later planting date allowed for warmer soil temperatures at planting, herbicides available for postemergence grass control in soybean were effective in reducing smooth brome and grass weed pressure, and soybean compensating ability under uneven stands is better than corn. Because yields were not reduced under NT, we recommend planting smooth brome CRP land to no-till drilled soybean the first year into crop production. Producers concerned about erosion should consider no-till planting of soybean on CRP land in the initial cropping season.

Weed control proved much easier with soybean because there were postemergence grass herbicides available to control any escapes following the fall spraying. When the research was started, Roundup Ready (glyphosate-tolerant) soybean or corn was not available. This technology may allow better perennial vegetation control and may make growing no-till soybean even more suitable for first-year conversion from CRP land to crop production. Herbicide-resistant corn hybrids {glyphosate, glufosinate-ammonium [2-amino-4-(hydroxymethylphosphinyl) butanoic acid-monammonium salt], and imidazolinone resistance} may also benefit no-till cropping systems.


   ACKNOWLEDGMENTS
 
The CRP to Crops Team recognizes that this research could not have been accomplished without cooperation and support from a number of sources. First, we want to emphasize the active support of the landowner, Charles Paulsen. The CRP to Crops project could not have been initiated or followed to completion without the understanding and flexibility of Mr. Paulsen. We utilized a Citizen's Advisory Committee comprised of LeRoy Hoesing, Garry Anderson, Bryce Neidig, Daryl McGhee, Neal Pohlman, David Speidel, Kris Thorp, Blake Brown, Lance Nearman, Stan Staab, Charles Paulsen, Hart Vollars, and Kurt Rewinkle. Ron Cleveland, Communications Associate, provided design services for many field days and publicity programs. Pat Bathke provided secretarial support throughout the project. Resources were provided as financial support and in-kind donations from the following organizations and agencies: AgrEvo, Bayer Ag, Crop Production Trust Fund, DowElanco, Dupont Agricultural Products, HACCO, Layman Foundation, Logan Valley Implement, Lower Elkhorn Natural Resources District, Monsanto, Northrup King, Pioneer Hi-Bred International, Natural Resources Conservation Service of USDA, University of Nebraska Agricultural Research Division, and University of Nebraska Cooperative Extension. Invaluable technical advice was given by Paul Jasa. Excellent prepublication reviews were made by Dr. Richard Clark and Dr. Gordon Carriker. Finally, without the encouragement of retired District Director emeriti, Robert Fritschen, this project would not have been initiated. This paper has been assigned Journal Series no. 12857, Agricultural Research Division, University of Nebraska. Mention of brand names is for descriptive purposes only. Endorsement or exclusion is not intended or implied.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 





This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Web of Science (2)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Shapiro, C. A.
Right arrow Articles by Stroup, W. W.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Shapiro, C. A.
Right arrow Articles by Stroup, W. W.
Agricola
Right arrow Articles by Shapiro, C. A.
Right arrow Articles by Stroup, W. W.
Related Collections
Right arrow Other Crop Management
Right arrow Other Soil Management
Right arrow Tillage

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