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


     

Published online 8 January 2009
Published in Agron J 101:215-225 (2009)
DOI: 10.2134/agronj2008.0001x
© 2009 American Society of Agronomy
677 S. Segoe Rd., Madison, WI 53711 USA
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 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 Google Scholar
Google Scholar
Right arrow Articles by Coelho, B. R. B.
Right arrow Articles by White, P.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Coelho, B. R. B.
Right arrow Articles by White, P.
Agricola
Right arrow Articles by Coelho, B. R. B.
Right arrow Articles by White, P.
Related Collections
Right arrow Nitrogen
Right arrow Nutrients
Right arrow Nutrient Management
Right arrow Tillage
Right arrow Animal Waste

Zonejection: Conservation Tillage Manure Nutrient Delivery System

B. R. Ball Coelhoa,*, R. C. Royb, A. J. Bruina, A. Moreb and P. Whiteb

a Agric. & Agri-Food Canada, Southern Crop Protection & Food Res. Cent., 1391 Sandford St., London, ON, Canada N5V 4T3
b Agric. & Agri-Food Canada, Southern Crop Protection & Food Res. Cent., Delhi, ON, Canada, N4B 2W9

* Corresponding author (ballb{at}agr.gc.ca).


   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY AND CONCLUSIONS
 REFERENCES
 
Manure application in minimum till (MT) systems is a challenge worthy of attention because residue cover is a keystone for environmental protection. To develop a system combining zone tillage and manure application into one operation (zonejection), two experiments were conducted. In Exp. 1, liquid swine manure (LSM) was applied in fall or spring for two site years (A, B). In Exp. 2, LSM was zone-applied either all preplant (PP) or split between preplant and sidedress (SP) for three site years (C, D, E). In both experiments, dietrich (DMI), vibro shank (VS), or subsurface deposition (SSD) applied the LSM, corn (Zea mays L.) was seeded in the manured zone, and NO3–N movement was monitored. Nutrients were supplied by inorganic fertilizer (IF) in control treatments under conventional till (CT), no till (NT), and zone till (ZT). With fall-applied LSM, after a mild winter, more N was lost from the soil–plant system (i.e., 35 kg ha–1 soil NO3–N) than after a cold winter with snow cover (18 kg ha–1), and corn grain yield was reduced (by 1.2 Mg ha–1), even though supplemental fertilizer N was sidedressed. In Exp. 2, with LSM zoned all PP or SP, grain yield and N use efficiency were comparable to that with IF, except when double the crop N requirement was zoned all PP (Site D). Planting into a zone of concentrated LSM (3.4 S m–1) reduced grain yield when the LSM was injected by VS. With careful management, zonejection allows efficient utilization of manure nutrients while preserving residue cover.

Abbreviations: ANR, apparent N recovery • CT, conventional till • DMI, dietrich • IF, inorganic fertilizer • LSM, liquid swine manure • MT, minimum till • NT, no till • PHSN, post-harvest soil nitrate • PP, manure applied all preplant • PSNT, presidedress nitrate test • SD, sidedress • SP, manure split between preplant and sidedress • SSD, subsurface deposition • UAN, urea ammonium nitrate • VS, vibro shank • ZT, zone till

Received for publication July 8, 2008.
   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY AND CONCLUSIONS
 REFERENCES
 
BROADCASTING IS A COMMON METHOD of manure application, but leaves the amendment vulnerable to runoff, particularly when not incorporated or when spread on frozen ground or snow (Withers et al., 2003; Daverede et al., 2004). Even with immediate incorporation, the applied material is more susceptible to surface movement than if injected (Topp et al., 2008). The process of incorporation exposes soil to erosion, adding further risk of surface water contamination. Erosion is a major source of pollutants in rural surface waters, but can be eliminated by NT management (Phillips et al., 1980). Considering the enormous benefit to surface water quality with NT, for example reduced P runoff by up to 91% relative to CT (Andraski et al., 2003), residue management is a critical component for controlling surface contaminant movement. Unfortunately, NT systems are sometimes rejected or abandoned due to poor corn yields and/or incompatibility with manure application.

Zone tillage is an alternative to fall plowing on soils where corn performs poorly under NT. The seed bed is drier, warmer, and less dense than under NT (Pierce and Burpee, 1995; Beyaert et al., 2002), corn yields can be comparable to that under CT (Opoku et al., 1997), and residue cover is maintained. Other practices that may enhance yields under NT or MT include planter-placed fertilizer and injection of N below the soil surface (Reeves et al., 1986; Vetsch and Randall, 2000). Such placement may alleviate the nutrient stratification problems that occur under NT (Yin and Vyn, 2002).

In systems that utilize liquid manures, injectors can place nutrients below the surface while leaving varying amounts of residue cover, depending on design. As compared with surface application, injection can increase corn yields (Schmitt et al., 1995; Ball Coelho et al., 2005) and reduce runoff by up to 94% (Daverede et al., 2004). The equipment should be capable of injecting liquid volumes typical of application rates. A larger sweep operating at greater depth and lower speed reduces manure exposure (Rahman et al., 2005). For tiled land, the injector should provide good mixing and tillage to minimize risk of contaminant movement into drains following application (Ball Coelho et al., 2005).

We hypothesized: i) manure application and ZT could be performed in a one-pass operation (zonejection), ii) placement in the root zone would allow efficient utilization of manure nutrients and help overcome the limitations to NT, iii) performing zone tillage in fall would improve corn yields relative to spring zone tillage in response to soil physical changes such as over winter mellowing during freeze-thaw, but that iv) fall application of manure would increase the risk of NO3–N contamination of groundwater due to loss in percolating water before the subsequent crop is actively absorbing nutrients. Our objectives were to develop a system where manure is applied during zone tillage in one operation, residue cover is maintained, corn yield is optimized, and potential for NO3–N movement to groundwater is minimized.


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 SUMMARY AND CONCLUSIONS
 REFERENCES
 
Cultural Practices and Equipment Development
Liquid swine manure was applied in fall or spring in site years A and B of Exp. 1, and during the production year either PP or SP between preplant and sidedress (SD) (mid to late June) in site years C, D, and E of Exp. 2. Soils were Typic Hapludalfs (Brunisolic Gray Brown Luvisols) in southwestern Ontario (Table 1 ). Sites A to D were silt loams of the Huron association in different fields of a commercial operation and previously cropped with winter wheat (Triticum aestivum L.). Site E was Embro series loamy glacial till and previously cropped with rye (Secale cereale L.). Small grains rotation was selected because it is common in eastern Canada and can cause a yield drag for the subsequent corn crop under NT management (Opoku et al., 1997).


View this table:
[in this window]
[in a new window]

  		Table 1. Selected characteristics in the top 20 cm of soil before imposing treatments at five site years of two experiments to develop a zonejection manure application system.

 
Three applicator systems were developed. The VS system had VS injectors (Kongskilde Ltd., Strathroy, ON) with 11-cm sweeps positioned between a coulter in front and covering disc behind. The DMI system was adapted from a custom-built zone tillage implement that had tiger points with 25-cm sweeps on NT shanks (Dietrich Machine Inc., Goodfield, IL) mounted between a cutting coulter in front and adjustable fluted coulters behind. This design was tested because of desirable mixing and tillage action and demonstrated capacity for corn yields comparable to CT (Beyaert et al., 2002). The third implement used to apply LSM was an AerWay SSD slurry applicator (Holland Equip. Ltd, Norwich, ON) that included a chopper distributor (Vogelsang, Essen, Germany). This low-disturbance equipment bands manure over aeration slots and was of interest because of potential for reduced runoff, NH3 emissions, and contaminant movement to tile drains relative to surface-broadcast or banding (Bittman et al., 2005; van Vliet et al., 2006; Lapen et al., 2008).

The VS toolbar was mounted behind a plot-scale applicator (2.65 m3 capacity) equipped with in-tank mixing (Nuhn Industries, Sebringville, ON), and an electromagnetic flow meter (Krohne Inc., Peabody, MA) and console (Raven Industries, Sioux Falls, ND) to facilitate calibration. For Exp. 1, the DMI and SSD ran in tandem with the tanker, which tracked alongside in borders to accommodate this traffic, using an umbilical line to transfer LSM to the implements. For Exp. 2, all three implements were developed into quick-hitch toolbars that mounted behind the tanker. This facilitated same-day completion of all treatments and eliminated the requirement to run the DMI and SSD alongside the tanker. For Exp. 1, a commercial model SSD (broadcast) was used that had 18-cm spacing between hoses that dropped material behind the 16 rolling tines. For Exp. 2, a new zone SSD was constructed having four double gangs of leaf style tines on 75-cm centers and eight hoses (rear drop), to till and apply LSM in zones similarly to VS and DMI. Injection depth was about 15 cm for VS and DMI at both experiments.

The dual purpose of the covering discs with VS, and rear coulters with DMI, was to cover the injected LSM with soil to minimize volatilization and runoff, and to create a raised seed bed, warmer and drier than the surrounding soil. Berm preparation for DMI and VS was performed in a separate pass at Site C, using adjustable fluted coulters mounted on a stand-alone toolbar. In subsequent site years (D, E), a toolbar with two adjustable offset wavy coulters was added behind the DMI and VS bars rather than hilling in a separate operation, because occasionally the hiller did not center over the LSM injection zone and the planter slid into the trench cut by the sides of the hillers.

To fill the applicator, a 14-m3 nurse tank was filled from a lagoon at a farrow-to-finish barn and transported to the experimental site for each LSM application. Both the nurse and applicator tanks were agitated during transport and application. For Exp. 1, the rate was 44 m3 ha–1 for all LSM applications. For Exp. 2, rates for the PP treatment were 56, 50, and 48 m3 ha–1, and for the SP were 19, 13, and 16 m3 ha–1 preplant plus 37, 36, and 45 m3 ha–1 SD, at Sites C, D, and E, respectively (Table 2 ). Rates were based on conductivity quick tests of the LSM before application and intent to supply crop N requirement entirely from LSM for Exp. 2. Samples of LSM from different loads on each application date were frozen until analyzed (Table 2). Total N concentration and LSM-N amount applied PP at Site D was unintentionally 1.5 to 2 fold greater than other times (Tables 2, 3 ), which provided opportunity to test the system response to over-application.


View this table:
[in this window]
[in a new window]

  		Table 2. Dates of selected sampling events (such as presidedress nitrate test [PSNT]) and cultural practices (such as urea ammonium nitrate [UAN] sidedress to all treatments of Exp. 1 and inorganic fertilizer treatments of Exp. 2, liquid swine manure application [LSM] in fall and spring [spr] for Exp. 1, and preplant [PP] and sidedress [SD] for Exp. 2) and composition of applied LSM.

 

View this table:
[in this window]
[in a new window]

  		Table 3. Nitrogen, phosphate, and potash applied as either inorganic fertilizer (IF) with the planter and sidedress (SD), or as liquid swine manure (LSM) using dietrich (DMI), vibro shank (VS), or subsurface deposition (SSD), in fall or spring for Exp. 1 (Sites A and B), and all preplant (PP) or split (SP) between preplant and sidedress for Exp. 2 (Sites C–E).

 
Both experiments included treatments with nutrients supplied by IF under CT, NT (benchmarks), or tilled using each implement without LSM (DMI IF, VS IF, and SSD IF) to differentiate physical from nutrient effects (Table 3). For Exp. 2, a CT control with no nutrient addition (CT0) was included to calculate apparent N recovery (ANR) and facilitate interpretation of soil N data. The CT plots were moldboard plowed on 19, 14, 30, 17 Nov. 2001, 2002, 2004, and 2005 at Sites A, B, D, E, respectively, and disked (usually two passes) on 21, 20, 19, 10, and 8 May 2002, 2003, 2004, 2005, and 2006 at Sites A to E, respectively. Site C was inadvertently fall chisel plowed before treatment establishment, so the NT benchmark was actually MT. Treatments were arranged in a randomized block design with four replications. Plots were 3 m wide (four corn rows) by 17 (Site E) to 20 (Sites A–D) m long.

Corn was planted in 75-cm rows at 74,100 seeds ha–1 using a John Deere 4-row planter equipped with a three- (Sites A, B) or single- (Sites C, D, E) coulter system, Keeton seed firmers (Precision Planting, Tremont, IL) and trash whippers (only used in NT for Exp. 2) (Table 2). Corn hybrid was DeKalb C42–71 each year except Site C (DeKalb C38–71). Seed was placed in the DMI and VS tilled zones in all years and in the SSD tilled zone for Exp. 2. All treatments received 10 kg ha–1 granular mini-MAP (mono-ammonium phosphate, NH4H2PO4) in-furrow. Nonmanured treatments, except CT0, received granular starter fertilizer banded 5 cm to the side and below the seed (Table 3). Weeds were controlled with glyphosate (N-[phosphonomethyl] glycine isopropylamine salt). For Exp. 1, all treatments were sidedressed with urea ammonium nitrate (UAN; [NH2]2CO NH4NO3) by knife-injection (Table 2) at varied rates (Table 3) depending on soil inorganic N presidedress (described below), so that differences in N availability would not confound evaluation of the systems. For Exp. 2, only IF treatments were sidedressed with UAN.

Soil, Water, and Plant Sampling and Analyses
Rainfall was measured by tipping buckets (TE525, Campbell Scientific Inc., Edmonton, AB). Soil water content in the top 30 cm was logged hourly at all sites using calibrated reflectometers (CS-615, Campbell Scientific Inc., Edmonton, AB). Additional reflectometers were installed 70 to 100 cm deep at Site E to determine water content near the depth of soil solution collection. Due to dry conditions during tasselling at Site D, corn was irrigated to minimize drought-related treatment bias. Drip lines with 30-cm emitter spacing were placed between the middle two rows of each plot. The 5-h irrigation event on 15 July 2005 increased soil water content by approximately 0.17 m3 m–3 (from 0.19 to 0.35 m3 m–3) in the drip-affected zone (Fig. 1B ).


Figure 1
View larger version (22K):
[in this window]
[in a new window]

  		Fig. 1. Daily average soil water content (when not frozen) at A) Sites A (2001–2002) and B (2003) of Exp. 1, and B) Sites C (2004), D (2005), and E (2006–2007) of Exp. 2, integrated over 0 to 30 cm except where indicated, average of two (Site A, B, C) or four (Sites D, E) reflectometers. Average of eight reflectometers at 70 to 100 cm deep, Site E.

 
Early corn growth was quantified after treatment differences were observed the first year (Site A). Corn shoots (8 at Site B, 16 at Sites C–E) were cut at the soil surface outside grain yield measurement areas (Table 2), rinsed, and ovendried (65°C). Shoot dry weight was converted to a hectare basis using June population count from the respective plot. Corn grain was harvested from the center two rows of each plot using a Winterstieger (Salt Lake City, UT) plot combine at Sites A to D and manually at Site E (Table 2) due to wet conditions. A representative grain subsample was retained for moisture (Dickey-John II, Auburn, IL), test weight, and N determination. Grain yield adjusted to 155 g kg–1 moisture was calculated from area combined. To determine stover N uptake, eight plants per plot outside grain yield measurement areas were cut at the soil surface at Sites D (28 Sept. 2005) and E (2 Nov. 2006). Cobs were removed and the stover was ovendried. Stover dry weight was converted to an area basis using stalk counts from the center two rows. To determine mature stalk NO3–N concentration, eight sections 15 to 35 cm above the soil surface were collected from each plot at Sites B to E within 2 d of harvest.

All plant samples were ground to <2 mm using a Wiley mill. Except for the grain and stover from Site E, the plant samples were subsequently ground finer by rolling on a conveyor in glass jars containing stainless steel rods, and combusted (LECO Corp., St. Joseph, MI) to determine total N. Grain and stover from site E did not require a finer grind, as total N was determined by acid digestion (Thomas et al., 1967), which accommodates a larger sample size (0.25 g) than combustion. Samples for the stalk nitrate test, a diagnostic of N management (Brouder et al., 2000), were also finely ground, 0.5 g (ovendry) shaken in 20 mL of deionized water for 1 h, and the extract filtered using Whatman no. 5. Agronomic efficiency was calculated based on grain yield (kg grain ha–1 / kg total N ha–1 in applied LSM or IF). Manure or fertilizer-N recovery was calculated based on N removed in harvested grain (kg grain N uptake ha–1 / kg total N ha–1 in applied LSM or IF). Apparent N recovery for Exp. 2 was calculated based on N removed in harvested grain (kg grain N uptake ha–1 [amended – check] / kg total N ha–1 in applied LSM or IF).

To determine soil inorganic N, samples were collected from the top 30 cm presidedress (presidedress nitrate test [PSNT]; Ontario Ministry of Agriculture and Food, 1999) and the top 20 cm post-harvest (post-harvest soil nitrate; PHSN) in all plots (2-cm-diam. core), and from deeper layers of the profile (5.1-cm-diam. core) periodically in selected treatments. Before SD, a composite of nine soil cores per plot were collected (Table 2), three of the cores positioned within the injection band and the other six between injection zones in the VS and DMI treatments. Within 1 wk of grain harvest, a composite of eight cores per plot was collected mid-row. Profile samples were obtained using a hydraulic apparatus (Ball Coelho et al., 2005) at: Site A in April 2002 from NT, CT, fall-, and spring-LSM and in November 2002 from NT, CT, fall-, and spring-LSM; Site B in April 2003 and November 2003 from NT, CT, fall-, and spring-LSM; Site C in July 2004 from all but ZT IF; Site D in fall 2005 from all treatments; and Site E in June 2007 from DMI PP, DMI SP, CT IF, NT IF, and CT0. Deep cores were positioned 20 cm from the injection zone if LSM had been applied. From Sites A, B, C, and E, cores were divided into five increments (0–20, 20–40, 40–60, 60–90, and 90–120 cm), and from Site D into seven increments (the five just mentioned, plus 120–150 and 150–180 cm). Soil samples were stored at 4°C and within 1 wk homogenized by pushing through a 6-mm-opening screen, then extracted by shaking 12.5 g field-moist soil in 25 mL of 2 M KCl for 1 h (Maynard and Kalra, 1993). Gravimetric water content was determined separately for each sample to correct concentrations to a dry weight basis. To monitor NO3–N movement continuously at Site E, two suction lysimeters per plot in DMI SP, DMI PP, CT IF, and CT0 treatments were installed using methods described by Ball Coelho and Roy (1997) with ceramic cups 120 cm deep. Soil solutions were collected weekly or following significant rainfall events (38 occasions from May 2006–June 2007).

Flow injection (Lachat Instruments, Milwaukee, WI) colorimetry was used to determine concentrations of NO3–N in soil and stalk extracts and in soil solutions (Diamond, 2001), and NH4–N in soil extracts and plant digests (Liao, 1999; Hofer, 2003). Soil NH4–N concentrations were measured in addition to NO3–N because the PSNT may not identify N responsive areas in manured soils (Katsvairo et al., 2003) and soil NH4–N can accumulate following preplant LSM application (Ball Coelho et al., 2004). Concentrations were however <1 mg NH4–N kg–1 for all but presidedress soil with spring-applied LSM at Site A (2.3 mg kg–1) and PP LSM in Exp. 2 (9 mg kg–1), so soil NH4–N data are not presented.

Data were analyzed according to the randomized complete block design using the MIXED procedure (SAS Institute, 1999). Data from Exp. 1 were analyzed with block specified as random, separately for Sites A and B due to variable weather that resulted in site-year interactions. These data were further subset for three balanced ANOVA to avoid confounding time of tillage with season of nutrient application. One subset tested season (fall, spring) and method (DMI, VS, SSD) of LSM application by comparing only manured treatments as a 2 x 3 factorial. A second subset tested nutrient source (fertilizer or LSM) within tillage type (NT, CT, MT [DMI, VS, SSD within type]) by comparing only spring-amended treatments. A third subset tested physical effects of time of tillage (fall, spring, none) within tillage type (NT, MT, CT) by comparing only IF treatments. For Exp. 2 data, site and block (nested within site) were specified as random and treatments were specified according to type (CT, ZT, NT, no nutrients), tillage (CT, ZT, NT), and source (LSM all preplant, LSM split, IF, no nutrients). Tillage type was designated MT for Exp. 1 and ZT for Exp. 2 because the SSD was a broadcast implement for Exp. 1 and zone for Exp. 2. Crop parameters from Sites C and E were analyzed separately from Site D due to the over-application in the PP treatment at Site D.

Profile soil NO3–N concentrations were analyzed separately for each depth increment, date, and site. Profile NO3–N amounts were calculated using soil bulk density = 1.34 Mg m–3 in the top 20 cm and 1.40 Mg m–3 below, as determined from undisturbed cores (0–40 cm) and soil weights in known core volumes collected below 40 cm deep. The MIXED procedure for repeated measures (SAS Institute, 1999) was used to analyze soil solution NO3–N data, with the 38 sampling events over time modeled using a repeated statement, treatment (nested within replicate) specified as the subject, and Akaike criteria to determine the appropriate structure of the covariance matrix (Littell et al., 1998). Normalization sometimes required log10 (PSNT Exp. 2; PHSN Site A and Exp. 2; NO3–N concentrations in soil 0–20 cm November 2002 Site A and 0–90 cm 2005 Site D; and in solutions Site E) or inverse (soil NO3–N concentration 40–60 cm Site C) transformation. When effects were significant, means were compared using the protected LSD at a 0.05 probability level (SAS Institute, 1999).


   RESULTS AND DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
SUMMARY AND CONCLUSIONS
 REFERENCES
 
Exp. 1: Fall and Spring Zonejection
LSM Season (Fall vs. Spring) and Method (DMI, VS, SSD), Nutrient Source (Spring LSM vs. Inorganic Fertilizer), and Physical Disturbance Effects
The effect of season of LSM application on corn early growth and grain yield was inconsistent. With spring-applied LSM at Site A, early corn growth was visibly improved relative to fall-LSM and comparable to CT. In contrast, shoot growth at Site B (quantified in July), was not affected by LSM application timing (Table 4 ). Spring-LSM as compared with fall-LSM improved grain yield, N uptake, and quality (test weight 712 > 693 kg m–3, moisture 200 < 220 g kg–1) at Site A, even though the fall-LSM treatment received additional sidedress N (50 kg ha–1 > spring-LSM, Table 3) to compensate for N losses over winter indicated by the PSNT (Tables 4, 5). At Site B, grain yield and N uptake were greater with fall- than spring-LSM (Tables 4, 5). Nitrogen use efficiency (agronomic efficiency, N recovery) was greater with spring- than fall-LSM at both sites (Table 5 ).


View this table:
[in this window]
[in a new window]

  		Table 4. Comparison, based on ANOVA, of soil NO3–N presidedress (PSNT) and postharvest (PHSN) and corn parameters as affected by season and method of liquid swine manure (LSM) application; nutrient source applied in spring; and physical disturbance type and season (fall or spring), for Exp. 1.{dagger}

 

View this table:
[in this window]
[in a new window]

  		Table 5. Effect of liquid swine manure (LSM) application season (fall or spring) and implement (dietrich [DMI], vibro shank [VS] or subsurface deposition [SSD]), spring-applied nutrient source (LSM or inorganic fertilizer), and tillage (conventional-[CT], minimum-[MT], no-till [NT]), on soil NO3–N presidedress (PSNT) and post-harvest (PHSN), and corn parameters for Exp. 1.

 
Method of LSM application had only minor effects on the crop and on soil NO3–N. At Site A, with SSD we observed uneven early corn growth (spring-LSM), and poorer grain quality (690 kg m–3 test weight, 220 g kg–1 moisture) than with the DMI injector (710 kg m–3, 200 g kg–1). These responses indicated that some LSM-N may have volatilized when applied by SSD, and so tine offset angle was set more aggressively in the second year of Exp. 1. Increasing tine angle reduces amount of residue remaining on the soil surface but improves incorporation. This may have subsequently reduced volatilization loss since PSNT at Site B tended to be greater for fall-LSM via SSD (19 mg NO3–N kg–1) than DMI (14 mg kg–1) or VS (13 mg kg–1). The fall-LSM SSD treatment thus received less inorganic fertilizer N SD and had greater agronomic efficiency and LSM-N recovery than VS or DMI (Tables 3, 5).

Inorganic fertilizer and spring-applied LSM resulted in similar grain yield and N uptake, although agronomic efficiency and N recovery were less with spring-LSM, particularly as compared with CT (Tables 4, 5). The addition of N fertilizer to spring-LSM amended corn may have reduced N use efficiency, albeit less was sidedressed than to the IF treatments due to greater PSNT (both sites, Tables 3–5GoGo).

Tillage did not affect corn yield at either site. Under CT, there were indications of more available N than under other tillage regimes, for example, greater PSNT and greater stalk NO3–N concentration at Site B; and there were positive responses to CT in some crop parameters such as lower grain moisture (190 < 200 g kg–1), greater test weight (730 > 710 kg m–3), N uptake, agronomic efficiency, and N recovery at Site A; greater July shoot weight (1.4 > 0.9 Mg ha–1), N uptake (47 > 32 kg ha–1), and agronomic efficiency and N recovery at Site B (Tables 4, 5). With DMI, there were positive early growth responses at Site B (averaged over spring-LSM and IF), for example, greater July shoot weight (1.0 Mg ha–1) than SSD (0.9 Mg ha–1), and greater shoot N uptake (42 kg ha–1) than VS (31 kg ha–1) or SSD (30 kg ha–1). Better early growth with DMI may be a response to the more aggressive tillage action of this implement than the VS or SSD.

Soil Profile NO3–N and N Balance
Soil NO3–N concentrations and corn early growth, grain yield, and N uptake all indicated that more N was lost from fall-applied LSM at Site A than B. The PSNT was greater with spring- than fall-LSM at both sites (Tables 4, 5), but the difference was larger at Site A: 46 (spring-) – 11 (fall-LSM) = 35 kg NO3–N ha–1, than Site B: 74 (spring-) – 56 (fall-LSM) = 18 kg NO3–N ha–1, indicating greater over-winter loss from fall-LSM at Site A. Soil profile NO3–N concentrations in the spring after fall LSM application were low at Site A and unaffected by the LSM, but at Site B, a pulse of NO3–N from fall-LSM was present 20 to 60 cm deep (Table 6 ). The pulse contained approximately 30 kg ha–1 of LSM-associated NO3–N (i.e., greater than comparable soil layers where no amendment was applied) and moved deeper (60–120 cm) by fall 2003 (Table 6). Season of LSM application did not affect residual NO3–N in the top 20 cm of soil post-harvest at Site A, but at Site B, topsoil in the fall-LSM treatment contained about 39 kg ha–1 more NO3–N than the spring-LSM treatment (Tables 4, 5), also indicating greater conservation of N from fall-LSM in the soil–plant system at Site B. Corn grain N uptake was 17 kg ha–1 less with fall- than spring-LSM at Site A, but 11 kg N ha–1 more with fall- than spring-LSM at Site B (Table 5).


View this table:
[in this window]
[in a new window]

  		Table 6. Comparison, based on ANOVA, of soil profile NO3–N concentrations in April with fall application of liquid swine manure (Fall LSM), fall plowing (CT), or untreated (NT), and in November with Fall or spring LSM or inorganic fertilizer under conventional or no-till management for Exp. 1.

 
Consistently colder, snowier winter months (November–February) and fewer potential thaw events (air temperature above 0°C) occurred at Site B (10 instances) than A (17 instances) between 1 December and 30 April (Fig. 2 A, B ). Soil profile NO3–N data provided no evidence that fall-applied LSM-N had leached below the corn root zone. Emissions of N2O are often greatest during freeze-thaw events (Wagner-Riddle and Thurtell, 1998; van Bochove et al., 2000), and so may have contributed to the N loss over the 2001–2002 winter. It is not known, however, whether denitrification in winter constitutes a significant sink for N (Phillips, 2008). Others have noted that fall manure application is inefficient in humid climates (Cote et al., 1999). For example, most of the N from fall-applied liquid (dairy) manure was lost over winter in a New York study (Katsvairo et al., 2003). Similarly, fall application of N fertilizer is not recommended in humid temperate regions unless a nitrification inhibitor is used (Randall et al., 2003).


Figure 2
View larger version (62K):
[in this window]
[in a new window]

  		Fig. 2. Monthly precipitation, average air temperature, and snow depth at (A) Site A, and (B) Site B of Exp. 1 with liquid swine manure zonejected in fall or spring. Snow depth and rain during fall 2001 were obtained from the nearest Environment Canada weather stations.

 
Exp. 2: Zonejection in Spring with LSM All Preplant or Split
Given no yield benefit to physical disturbance in fall, contrary to our expectations based on previous work (Kwaw-Mensah and Al-Kaisi, 2006; Perez-Bidegain et al., 2007), and variable yield response and poor N-use efficiency with LSM application in fall, subsequent investigation focused on LSM zonejection in the production year.

Nutrient Source (LSM vs. Inorganic Fertilizer), Time of LSM Application (Preplant vs. Split), and Physical Disturbance Effects
With LSM zonejected in season at rates balanced with crop N demand, corn yield and N use efficiency were comparable or improved relative to that of corn fertilized with inorganic nutrients (Tables 7 , 8 ). Agronomic efficiency and N recovery, respectively, were 71 kg grain ha–1 (kg N applied ha–1)–1 and 0.85 kg grain N uptake ha–1 (kg N applied ha–1)–1 for all treatments in site years C and E. The ANR averaged 0.35 at Site C and 0.24 at Site E, comparable to values reported in other studies for both inorganic fertilizer and sidedressed LSM (Cassman et al., 2002; Ball Coelho et al., 2006).


View this table:
[in this window]
[in a new window]

  		Table 7. Comparison, based on ANOVA, of corn parameters as affected by tillage (conventional-, zone-, no-till) and nutrient source (liquid swine manure [LSM] all preplant, or split between preplant and sidedress, inorganic fertilizer, no nutrients) at Site D with 2 x LSM-N preplant, and Sites C and E of Exp. 2.

 

View this table:
[in this window]
[in a new window]

  		Table 8. Effects of tillage (conventional- [CT], no- [NT], or zone-till [ZT]) and nutrients (none [CT0], inorganic fertilizer [IF], or liquid swine manure [LSM] applied either all preplant [PP] or split between preplant and sidedress [SP]) on soil NO3–N presidedress (PSNT) and corn parameters at Site D with 2x LSM-N preplant, and Sites C and E of Exp. 2.

 
When double the LSM-N rate was zoned PP (Site D), some corn parameters were negatively affected (Table 9 ). Over-application with the VS applicator was the only case where grain yielded less with LSM than IF. The 2x LSM-N application reduced June population in PP (74,000 plants ha–1) relative to SP and IF (77,000 plants ha–1), and suppressed early corn growth (dry weight, N uptake) where applied by VS and DMI, but not SSD (Table 9). Germination and early growth may have been affected by salt as opposed to NH4–N toxicity. The NH4–N concentrations and amounts applied PP at Site D were comparable to LSM applied at other sites, whereas the LSM had highest conductivity (Tables 2, 3), nearly 10-fold the salt tolerance threshold level for corn of 0.37 S m–1 in soil water (Hoffman et al., 1983). The SSD applicator likely mixed the material into a greater volume of soil, reducing osmotic potential around the emerging seedling. Nitrogen accumulated in stover with the over-application, evident as increased (relative to CT) stalk NO3–N concentrations with all three methods of application, increased stover N concentration and uptake with DMI and VS, and increased stover dry weight with VS (Table 9). Grain N was unaffected and so grain-based measures of N use efficiency (agronomic efficiency, N recovery, ANR) were reduced with the 2x N rate PP at Site D relative to IF or SP (Table 8). Sequestering N in corn stover would render it less available for NO3–N leaching. This type of unexpected feedback response by the system is therefore an important consideration for risk assessment modeling.


View this table:
[in this window]
[in a new window]

  		Table 9. Tillage (conventional- [CT], no- [NT], or zone till via dietrich [DMI], vibro shank [VS], or subsurface deposition [SSD]), nutrient source (inorganic fertilizer [IF], liquid swine manure [LSM], or none [CT0]), and timing of LSM application (all preplant [PP] or split between preplant and sidedress [SP]) effects on corn parameters at Site D with 2x LSM-N preplant, and Sites C and E of Exp. 2.

 
Application equipment and timing had minimal effects on the crop other than at the 2x LSM-N rate, which was the only instance (VS Site D) that split-application of LSM provided a yield advantage relative to the all preplant treatment. A timing response was noted at Site C with greater early shoot N concentration and N uptake in all PP than in split-applied LSM (Table 8). Positioning hoses in front of the SSD tines instead of behind was tested adjacent to Site C to determine whether N volatilization losses with rear manure drop reduce crop yield. With front drop, corn yielded similarly (10.2 Mg ha–1) to the Site C SSD treatment average (10.6 Mg ha–1), so hose configuration behind tines was continued in subsequent site years of Exp. 2.

Corn grain yield was not responsive to tillage at Exp. 2 (i.e., CT = NT; Tables 8, 9). Other crop parameters indicated some benefit to the physical disturbance of CT or ZT; for example, grain N concentration (Sites C, E) and total N uptake (Site E) were greater under CT than NT; total N uptake with VS and DMI (averaged over LSM and IF, Site E) and most mature crop parameters with DMI (Sites C, E) were comparable to CT and greater than NT (Table 9).

Soil Profile and Soil Solution NO3–N and N Balance
There was no evidence from the Exp. 2 soil profile or soil solution samples that the zonejection system increased risk of NO3–N movement to groundwater relative to supplying nutrients from inorganic fertilizer. At Site C, even after 145 mm of rain between preplant application and sampling mid-July (Fig. 3A ), N from PP LSM was detected (as soil NO3–N > IF) only in the top 40 cm (Table 10 ). In the fall (2004), PHSN accounted for <5 kg NO3–N ha–1 and did not vary with treatment (ANOVA not shown), indicating that the LSM N had been taken up by the corn crop. This was expected since corn grain contained more N than was applied to the PP and IF treatments (Tables 3, 8).


Figure 3
View larger version (26K):
[in this window]
[in a new window]

  		Fig. 3. Monthly (A) precipitation, and (B) average air temperature at Sites C, D, and E of Exp. 2 with liquid swine manure application preplant and sidedress.

 

View this table:
[in this window]
[in a new window]

  		Table 10. Comparison, based on ANOVA, of soil profile NO3–N concentrations with conventional- (CT), no- (NT), or zone till (ZT) and liquid swine manure (LSM) applied all preplant (PP) or split (SP), inorganic fertilizer (IF), or no nutrients (CT0) for Exp. 2.

 
At Site D, soil profile NO3–N concentrations in the fall were greater with the 2x LSM-N rate (PP) than in SP or IF from 20 to 90 cm deep, particularly in the 20- to 40-cm layer (Table 10). Nitrogen also moved down in SP and IF treatments, as indicated by greater NO3–N concentrations at 20 to 90 cm in SP, and at 60 to 90 cm in IF than in CT0 (Table 10). There was 84 kg ha–1 more soil profile NO3–N with PP LSM than CT0, similar to the amount by which total N application in PP exceeded corn aboveground uptake (329 kg LSM-N ha–1 applied – 244 kg N ha–1 uptake = 85 kg N ha–1). Profile amounts were 48 and 28 kg ha–1 more NO3–N than CT0 in SP and IF treatments, respectively, where aboveground (grain + stover) N uptake exceeded N applied by about 55 kg ha–1. There was no accumulation of NO3–N in topsoil after harvest (PHSN) with the 2x rate PP, probably because N had moved to the 20- 40-cm layer and had accumulated in stover (Tables 9, 10).

At Site E, NO3–N concentrations in weekly suction lysimeter samples (P [treatment] = 0.48), in topsoil fall 2006 (PHSN), and in the profile the following spring (2007), were all equal regardless of whether LSM, IF, or no nutrients had been applied. Soil concentrations on 4 June 2007 averaged 3.2, 1.6, 1.2, 0.8, and 0.6 mg NO3–N kg–1 at 0 to 20, 20 to 40, 40 to 60, 60 to 90, and 90 to 120 cm, respectively. The concentration at 90 to 120 cm deep converted to a solution basis using soil water content of 0.3 m3 m–3 equaled 2 mg NO3–N L–1 solution, confirming reliability of measures from lysimeters, which also averaged 2 mg NO3–N L–1 in June 2007. Corn aboveground uptake was greater in IF, equivalent in PP, and about 25 kg ha–1 less in SP than the amount of N applied. About 11 kg NO3–N ha–1 moved below 120 cm throughout 2006 and into the following spring, based on average (of CT IF, CT0, PP LSM, SP LSM) solution NO3–N concentrations and rainfall between sampling events when the soil profile was saturated (>0.40 m3 water m–3 in topsoil, >0.38 m3 water m–3 in subsoil, Fig. 1B).


   SUMMARY AND CONCLUSIONS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY AND CONCLUSIONS
REFERENCES
 
With zonejection in fall, LSM-N availability and crop yield response were variable. After a rainy, mild winter, proportionately less NO3–N remained from fall-applied LSM (24% of the soil NO3–N with spring-LSM) than after a cold winter (75% remaining), and corn yields were reduced, despite compensational application of sidedress N fertilizer. Crop N use efficiency (agronomic efficiency and N recovery) was less with fall- than spring-LSM, and less by corn amended with LSM in either fall or spring (which were both sidedressed with supplemental fertilizer N) than by corn supplied exclusively with inorganic fertilizer nutrients under CT.

When LSM was zonejected during the crop year (preplant) and no additional N fertilizer was added (Exp. 2), grain yields, agronomic efficiency, and N recovery were comparable to corn grown under CT with inorganic fertilizer, and crop aboveground N uptake was similar to the amount of LSM-N applied (Sites C and E). Both the all preplant and split LSM application treatments resulted in high agronomic efficiency and N recovery, except in the case where LSM was over-applied (PP Site D). With a doubled LSM-N rate, N use efficiency measures based on grain N uptake were less than for corn supplied with inorganic fertilizer nutrients, and a biologically significant N amount accumulated in the corn stover.

The VS, DMI, and SSD were all suitable implements for zone application. Differences in equipment were small with some evidence of better early growth with DMI, less risk of injury with SSD if highly concentrated manure is applied, and suitability for SD with VS due to narrow disturbance. The zonejection system provides opportunities to: significantly reduce fuel consumption as it replaces fall plowing, spring secondary tillage, and nutrient application; avoid surface nutrient stratification associated with NT; leave residues largely intact for runoff and erosion control; and manage odor. Additional equipment and labor requirements relative to broadcasting are offset by eliminating the incorporation pass, the seedbed preparation passes, and fertilizer inputs. The system is adaptable to other crops that are grown in wide rows and have high N requirements, such as sorghum, sunflower, sugar beets, black tobacco, potato, tomato, and cole crops. Zonejection, one-pass zone till using the injector to both place manure nutrients and form a seed bed, displayed low risk of NO3–N movement to groundwater in silt loams with no artificial drainage; however, sensitivity by the crop to over-application and efficient use of the N applied warrant careful attention to rate when using this system.


   ACKNOWLEDGMENTS
 
Ontario Pork, AAFC's Matching Investment Initiative, Monsanto Canada Inc./Dekalb: financial support; T. McDowell, R. Murray, R. Lester and K. Henning: technical support; G. Milliken: statistical advice; Bloxslea Farms, C. Brown, G. McGregor, Nuhn Industries, Aerway/Holland Equipment, Buckrell Soil & Nutrient Management, W. Graham, and AAFC farm crew: logistical support.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY AND CONCLUSIONS
 REFERENCES
 
All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY AND 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 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 Google Scholar
Google Scholar
Right arrow Articles by Coelho, B. R. B.
Right arrow Articles by White, P.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Coelho, B. R. B.
Right arrow Articles by White, P.
Agricola
Right arrow Articles by Coelho, B. R. B.
Right arrow Articles by White, P.
Related Collections
Right arrow Nitrogen
Right arrow Nutrients
Right arrow Nutrient Management
Right arrow Tillage
Right arrow Animal Waste

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