Impact
Statement

PDF version
for printing

© 2005 Plant Management Network.
Accepted for publication 23 November 2004. Published 20 January 2005.

Manure Nitrogen Management for Corn on Loess Soil in a Sensitive Groundwater Area

J. S. Strock, Southwest Research and Outreach Center, University of Minnesota, Lamberton 56152; and M. A. Schmitt, Department of Soil, Water, and Climate, University of Minnesota, St. Paul 55108

Corresponding author: J. S. Strock. jstrock@umn.edu

Strock, J. S., and Schmitt, M. A. 2005. Manure nitrogen management for corn on loess soil in a sensitive groundwater area. Online. Crop Management doi:10.1094/CM-2005-0120-01-RS.

Abstract

A number of community water supplies in southwest Minnesota are threatened by nitrate contamination. Manure nitrogen (N) management experiments were conducted in southwest Minnesota from 2000 to 2003 to compare manure management practices (timing, rate, and method of application) for preventing negative nitrate-nitrogen (NO₃-N) impacts on shallow ground water. Differences among manure rates produced a significant corn grain yield response in one out of four years. Mean soil inorganic N concentrations measured throughout the season showed significant soil NO₃-N differences for all manure rates in all years. Mean cornstalk NO₃-N concentrations also showed significant cornstalk NO₃-N differences for all manure rates in all years. When used to evaluate manure N management outcomes, sensible manure management practices along with N diagnostic testing can be excellent comprehensive management tools for livestock producers.

Introduction

Water quality is an important national concern among the general public and within the agricultural community. In many regions, important ground water aquifers that serve as a primary source of drinking water for rural communities, farms, and large cities underlie large areas of agriculturally important crop land. Livestock manure is a potential source of NO₃^- to ground water (3). Application of excessively high manure rates relative to crop needs has been shown to be the cause for high NO₃^- in ground water (5).

The guiding principle to minimize NO₃^- loss to ground water is to minimize the amount of NO₃^- in the rooting zone, especially during periods when leaching is likely to occur. Availability of N from manure is an important factor as are timing and method of application. Soil N testing is a useful management tool for improving N recommendations for corn. There are two distinctive approaches that emerged over the last three decades for soil N testing: the preplant soil N test and the in-season soil N test (6,8,11). Another management tool, the cornstalk NO₃^- test, is a relatively new test for measuring the sufficiency of N for corn. Recent studies have shown that the NO₃^- concentration in the lower portion of cornstalks at the end of the growing season may be used to assess the N status of the current corn crop and to evaluate nitrogen management planning for the coming year (1,2,4).

The potential impact of manure applications on shallow ground water is a concern for a number of community water supplies in southwest Minnesota that are threatened by NO₃^- contamination. Our objectives for this study were to compare manure management practices (timing, rate, and method of application) for preventing negative nitrate impacts on shallow ground water, and if necessary, refine, and modify Best Management Practices (BMPs) for nitrogen for cropping systems in this region.

Field Experiment Comparing Manure Management Practices

Manure application timing, rate, and placement experiments were conducted at 4 sites in southwest Minnesota from 2000 to 2003. Sites were located on farmer-cooperator fields and were selected to include soils types and crop management systems represented in the loess region of southwest Minnesota. No manure was applied within one and a half years before establishment of the experiment. Expected corn yield for this region of Minnesota typically range between 125 and 149 bu/acre. According to University of Minnesota guidelines, 90 lb of N per acre is required to produce at 125 to 149 bu of corn per acre on these medium to high organic matter soils following soybean (8).

Site characteristics and details of manure application, and manure characteristics for the sites are listed in Table 1 and Table 2. Soils in this region of southwest Minnesota developed in loess deposits over glacial till or sand and gravel and are generally well-drained. This 4-year study provided an opportunity for field evaluation of manure management in the northern Corn Belt under contrasting growing conditions. Precipitation, given on a calendar year basis, was highly variable during the study period (Table 3), including a below-average year (2003), two nearly normal years (2000 and 2002), and an above-average year (2001).

Table 1. Site characteristics and manure application information for 2000-2003.

Table 2. Manure analysis results for 1999-2002.

 † Values represent total N, P, and K.

Table 3. Mean monthly precipitation (inches) for the period 1999-2003

Treatments were located in 20-ft-by-50-ft plots and replicated four times in a randomized complete block design. At each site, manure was applied about October 1 and again about October 30. In addition to an unfertilized control plot (manure at 0 gal/acre), manure rates were 2500, 5000, and 7500 gal/acre of finishing swine (Sus scrofa domesticus) manure. Manure placement was either injected to a depth of 8 inches beneath the soil surface on 30-inch spacing, or broadcast on the soil surface without incorporation.

Grain yield was measured from each plot after physiologic maturity. Yield determination was made by harvesting four 40-ft long rows from the eight-row plots using a plot combine. Grain moisture content was determined and yields were calculated and expressed on a 15.5% moisture basis.

Soil test phosphorus (P) and potassium (K) were determined from 6-inch soil cores collected from unfertilized check plots in autumn. Soil tests at the four sites were "medium" to "high" Bray P₁ and "high" exchangeable K (8). Soil sampling was conducted for N testing purposes three times each year. Soil samples were collected from the control plots approximately 2 weeks before planting. Eight cores were collected and composited from the middle of each plot. A systematic sampling scheme was used that consisted of selecting two random areas and collecting four vertical soil cores. Soil cores were collected at 1-ft increments to a depth of 2 ft and at 7.5-inch intervals, beginning in the row, on a transect perpendicular to the direction of planting and possible manure application zones. Soil samples were collected from all plots in mid June, about 5 to 6 weeks after planting, when corn was approximately 12 inches tall (V4). These samples were collected according to the same sampling scheme as preplant samples but from a 1-ft depth. Soil samples were also collected from all plots in October post harvest. Three soil cores were collected from three random areas to a depth of 2 ft in 1-ft increments. Preplant and in-season soil samples were collected using hand probes and autumn samples were collected using a hydraulic probe.

Soil samples were dried at 95°F in forced-air driers, ground, and passed through a 0.08-inch (2-mm) sieve. Ammonium- and nitrate + nitrite-N were determined from using the colorimetric Cd-reduction method after 2 M KCl extraction.

Cornstalk testing was used to assess the N status of the corn crop at physiologic maturity by measuring the NO₃^- concentration in the lower portion of the cornstalk. Cornstalk samples were collected two to three weeks after physiological maturity. An 8-inch long segment of stalk was removed from between 6 and 14 inches above the soil. Ten cornstalk samples were collected from each plot. Cornstalk samples were dried at 140°F in a forced-air drier and ground. Nitrate-N was determined from using the colorimetric Cd-reduction method after 1 M KCl extraction.

Analyses of variance (ANOVA) was performed for each variable using the GLM procedure using SAS (10). All interactions and main effects of manure application timing, placement, and manure rate on corn grain yield and soil nitrate-N concentration were determined by averaging over years for each site. When the F-test from the ANOVA was significant for a treatment effect, a mean separation test (t-test) was calculated (P < 0.05).

Corn Grain Production

Corn yields showed no significant effect of time or method of application and no significant interactions, but year and manure N application rate were significant (Table 4). The amount of N supplied in the 2500 gal/acre manure rate supplied adequate available N such that N was not a limiting factor in three out of four years. Mean corn grain yields ranged from 70 bu/acre with no manure to 155 bu/acre based on year and manure application rate (Table 5). Analysis of liquid swine manure indicated adequate levels of P and K for corn production (8).

Table 4. Analysis of variance table (Pr. > F).

 † PP-NO₃ = preplant nitrate; PP-NH₄ = preplant ammonium; PS-NO₃ = presidedress nitrate; PS-NH₄ = presidedress ammonium; rsn-NO₃ = residual soil nitrate; rsn-NH₄ = residual soil ammonium; ST-NO₃ = stalk nitrate.

 * Less than 0.0001.

Table 5. Corn yield averaged across time and method of application from 2000-2003.

 ^† Average of early and late autumn total manure N.

 ‡ Means followed by the same letter with a column are not significantly different at the 0.05 probability level.

There was a statistically significant effect of manure N application rate on yield in 2000, 2002, and 2003; however, the effect was only between the control and the lowest application rate as all three manure N rates resulted in statistically similar yields. In 2000, corn yield means ranged from 146 bu/acre for the 2500 gal/acre manure treatment to 149 bu/acre for the 7500 gal/acre manure treatment. The 2500 gal/acre treatment resulted in a 29% (42 bu/acre) yield increase compared with the control. The lack of significant yield effects among the three manure application rates in 2000 was not surprising as the lack of autumn precipitation and below normal growing season precipitation greatly reduced the potential for N losses due to leaching. In 2001, there was a significant effect of manure application rate on yield between treatments (Table 5). Corn yield means ranged from 70 bu/acre for the control to 115 bu/acre for the 7500 gal/acre manure treatment. The 7500 gal/acre manure rate resulted in a 39% (45 bu/acre) yield increase compared with the control. Inferior yields in 2001 can be attributed to environmental stress from hot, dry, windy conditions during silking. During this period pollination occurs and the number of kernels is determined. Environmental stress caused by hot, dry conditions during the silk phase may reduce silk growth and develop or desiccate pollen resulting in poor pollination (9). In 2002, the 2500 gal/acre manure rate resulted in a 15% (23 bu/acre) yield increase compared with the control and in 2003 the 2500 gal/acre treatment resulted in a 21% (25 bu/acre) yield increase compared with the control. In 2002, dry conditions early in the growing season reduced the potential for N loss and warm temperatures coupled with near normal growing season precipitation resulted in excellent kernel development. In 2003, despite early season weather related stresses, plants pollinated well and resulted in good kernel development.

Soil Nitrogen

Soil N testing showed significant effects of rate and method of application and significant year by treatment interactions (Table 4). Lack of incorporation of broadcast manure leads to greater N loss through ammonia volitalization and consequently less measurable soil N, as evidenced by the significant effect of method of application. Mean soil inorganic N concentrations measured throughout the season showed significant soil NO₃-N differences for all treatments in all years and soil NH₄-N differences for all treatments in 2001-2003 (Table 6). Ammonium-N concentrations measured in 2000 were not affected by the treatments.

Table 6. Soil nitrate-N and ammonium-N concentration averaged across time and method of application for various sampling times and depths for 2000-2003.

 † Control (zero gal of manure) not included in statistical analysis.

 ‡ Means followed by the same letter with a column are not significantly different at the 0.05 probability level.

Soil NO₃-N concentrations were correlated to manure application rate, although this relationship was not linear (data not shown). Our data showed that soil NO₃-N concentrations increased as manure application rate increased for all sampling times (Table 6). For example, in 2000, in-season soil NO₃-N concentrations in the 0-to-1-ft soil depth of the control treatment had a mean soil NO₃-N concentration of 7.5 ppm NO₃-N. Because no manure was applied to this treatment, the NO₃-N present was either from carryover NO₃-N from the previous year or NO₃^- released from soil organic matter. The initial 2500 gal/acre application rate increased soil NO₃-N by 53% whereas the 5000 and 7500 gal/acre manure application rates increased NO₃-N by 58% and 64%, respectively.

Preplant soil NO₃-N concentrations measured from 2-ft soil depth showed significant soil NO₃-N differences for all treatments in all years (Table 6). Soil NO₃-N measured from a 2-ft depth in spring before corn planting has been used as a diagnostic tool to indicate the residual N credit, either from manure or commercial fertilizer (8,11). Soil N credit amounts are a function of the preplant soil NO₃-N concentration. At preplant soil NO₃-N concentrations between 12.1 and 15.0 ppm, 95 lb of N per acre is expected (8). Data in Table 6 indicated that the amount of available N, assuming 80% first-year N availability, supplied in the 2500 gal/acre manure rate was sufficient for optimal yield for all years except 2002 (Table 6). Early-season soil NO₃-N measurements also provide an indication that possible NO₃-N loss occurred. Since the N supplied by the 2500 gal/acre manure rate was sufficient for optimal yield in 3 out of 4 years, the additional N supplied at higher application rates was a potential environmental liability. Data in Table 6 do not provide any evidence of NO₃-N loss due to leaching except for 2002. Nitrate-N loss due to leaching during 2002 was unlikely because spring precipitation was generally below normal (Table 3). However, early season weather conditions probably limited mineralization of N, which resulted in lower than expected soil NO₃-N concentrations from preplant soil sampling.

In-season soil NO₃-N concentrations measured from 0-to-1-ft soil depth showed significant soil NO₃-N differences for all manure rates in all years. Soil NO₃-N measured from a 1-ft depth when corn is approximately 12 inches tall is a diagnostic tool used as an indicator of potentially mineralizable organic-N (6). Data in Table 6 indicated an increase of 40 to 53% in soil NO₃-N concentrations between preplant and in-season sampling for all manure rates in 2001. Favorable early season growing conditions resulted in release of N from soil organic matter, decaying plant residue, and decomposing manure.

Nitrate-N concentrations measured after grain harvest were relatively low compared with any early season concentrations (Table 6). Mean soil NO₃-N concentrations after grain harvest in 2001 were considerably higher, ranging from 6.8 to 21.5 ppm, than soil NO₃-N concentrations measured in other years. Weather conditions that limited grain yield during 2001 could explain why post-harvest soil NO₃-N concentrations were high. In addition, dry conditions like those experienced from June through December 2001 can result in a build up of soil NO₃-N in the soil profile that can later lead to substantial NO₃-N losses in subsequent wet years (7). Thin soils with shallow water tables like those in this study would be particularly vulnerable.

Cornstalk Nitrate

Cornstalk N testing showed significant effects of rate and method of application and significant year by treatment interactions (Table 4). Annual mean cornstalk NO₃-N concentrations showed significant differences among manure rates across all sites and years (Table 7). Cornstalk NO₃-N was highly variable during the study period but the general trend indicated that mean cornstalk NO₃-N increased as manure application rate increased. These results showed that weather conditions, soil N availability, and N uptake by the plant varied considerably among years. In three of four years (2000, 2001, and 2003) cornstalk NO₃-N concentrations greater than 1000 ppm indicated that 2500 gal/acre manure was sufficient for optimal yield according to the categories defined by Blackmer and Mallarino (2).

Table 7. Corn stalk NO₃^- concentration averaged across time and method of application for 2000-2003.

 † Control (zero gal of manure) not included in statistical analysis.

 ‡ Means followed by the same letter with a column are not significantly different at the 0.05 probability level.

During 2001, cornstalk NO₃-N concentrations were extremely high (Table 7). Extremely high in-season soil NO₃-N concentrations and high post-harvest residual soil NO₃-N concentrations (Table 6) indicated that N availability was not a limiting factor for corn grain yield. Weather conditions during reproductive stages and kernel development limited grain yield during 2001 and could explain why cornstalk NO₃-N concentrations and soil NO₃-N concentrations did not agree with manure rate for optimal yield. Similar weather and growing conditions existed in 2003 and would account for comparable cornstalk NO₃-N results. In 2002, mean cornstalk NO₃-N concentration (529 ppm) indicated that corn grain yield was potentially limited by N availability at the 2500 gal/acre manure rate (Table 7). According to Blackmer and Mallarino, cornstalk NO₃-N concentrations between 250 and 700 ppm indicate that N availability is close to optimal but could result in lower yields (2). Balkom et al. found that 62% of the year-to-year variation in mean cornstalk NO₃-N concentration was explained by early-season precipitation (1). They found that cornstalk NO₃-N concentration decreased as the amount of early season precipitation increased.

Conclusions

Field experiments were conducted to compare manure manage ment practices (timing, rate, and method of application) for optimizing yield and preventing negative nitrate impacts on shallow ground water. Although no significant effect of method of application on yield was found, method of application did affect preplant and in-season soil NO₃-N concentrations that influenced yield. Nitrogen application rate played a major role in influencing the amount of inorganic soil N during the growing season and cornstalk NO₃^- concentrations at the end of the growing season. Although these differences did not translate into additional yield they do translate into increased environmental liability.

In this study, increasing soil NO₃-N and end-of-season cornstalk NO₃-N concentrations along with increasing manure application rates were indicative of manure N management practices. When used to evaluate manure N management outcomes, sensible manure management practices along with N diagnostic testing can be excellent comprehensive management tools for livestock producers. It is clear from this study that N diagnostic test such as soil testing for inorganic N, especially during the growing season, as well as end-of-season cornstalk NO₃-N testing can contribute to an N management program. These tools provide an effective way to evaluate and improve manure N management practices. Limiting the potential for N loss and quantifying the sufficiency of soil N for optimum grain yield is possible for corn production in an environmentally sensitive area.

Acknowledgments

The authors wish to thank Mark Coulter, Steve Iverson, Andrew Scobbie, and Thor Sellie for their contributions to conduct of this project. Gratitude is extended to the Minnesota Department of Agriculture for their contributions to this project. Appreciation is also expressed to Lincoln Pipestone Rural Water Supply District and the Legislative Commission on Minnesota Resources for providing financial support of this project.

Literature Cited

1. Balkom, K. S., Blackmer, A. M., Hansen, D. J., Morris, T. F., and Mallarino, A. P. 2003. Testing soils and cornstalks to evaluate nitrogen management on the watershed scale. J. Environ. Qual. 32:1015-1024.

2. Blackmer, A. M., and Mallarino, A. P. 1996. Cornstalk testing to evaluate N management. Iowa State Univ. Ext. Serv. PM-1584.

3. Bouldin, D. R., Skausner, S. D., and Reid, W. S. 1984. Use of nitrogen from manure. Pages 663-676 in: Nitrogen in Crop Production. R. D. Hauk, ed.-in-chief. ASA-CSSA-SSSA, Madison, WI.

4. Brouder, S. M., Mengel, D. B., and Hofmann, B. S. Diagnostic efficiency of the blacklayer stalk nitrate and grain nitrogen tests for corn. Agron. J. 92:1236-1247.

5. Keeney, D. R. 1986. Sources of nitrate to ground water. CRC Crit. Rev. Environ. Control. 16:257-304.

6. Magdoff, F. 1991. Understanding the Magdoff pre-sidedress nitrate test for corn. J. Prod. Agric. 4:297-305.

7. Randall, G. W. 1998. Implications of dry and wet cycles on nitrate loss to subsurface tile drainage. Pages 53-60 in: Proc. 7th Annual Drainage Symposium. Drainage in the 21st Century. 8-10 March, 1998. Orlando, FL.

8. Rehm, G., Schmitt, M., Randall, G., Lamb, J., and Eliason, R. 2000. Fertilizing corn in Minnesota. Univ. of Minnesota Ext. FO-3790-C.

9. Ritchie, S. W., Hanaway, J. J., and Benson, G. O. 1997. How a corn plant develops. Iowa State Univ. of Sci. & Tech. & Coop. Ext. Service, Ames, IA.

10. SAS Institute. 1990. SAS/STAT user’s guide. SAS Inst., Cary, NC.

11. Schmitt, M. A., and Randall, G. W. 1994. Developing a soil nitrogen test for improved recommendations for corn. J. Prod. Agric. 7:328-334.