| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Ministry of Agric. (INTA) Dep. of Soil and Land Evaluation, Centro Colon, San Jose, Costa Rica
b Dep. of Agronomy, Iowa State Univ., Ames, IA 50011-1010
c emeritus, Dep. of Agric. and Biosys. Eng., Iowa State Univ., Ames, IA 50011-1010
* Corresponding author (jsawyer{at}iastate.edu)
Received for publication September 5, 2006.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: CM, SPAD chlorophyll meter • COD, chemical oxygen demand • TKN, total Kjeldahl nitrogen
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Estimating manure crop N availability, and therefore manure usage rate, is complicated because of uncertain potential availability of manure N (Killorn, 1998) and season-long crop N uptake requirements (Schepers et al., 1998). Binder et al. (1996) stated the importance of manure mineralization and synchronization of N release with crop demand. Manure characteristics, reactions in soil, and timing of N availability for crop use can change when manure is chemically or physically manipulated, as with anaerobic digestion (Sutton et al., 1978; Bernal and Kirchmann, 1992; Kirchmann and Lundvall, 1993), thus potentially requiring different manure management practices to gain best use of manure N for crop production. Improved estimates of crop response and N availability from manipulated manure sources will help farmers with manure management plans and application rate management. Swine manure is considered a valuable nutrient source that can increase yields when applied to soil at rates consistent with good agronomic practices (Duffera et al., 1999). This is attributed to improved soil physical properties and enhanced nutrient supply. The nutrient benefits, however, are not well defined with digested manure.
In the anaerobic digestion process to produce biogas, bacteria break down organic matter in an oxygen-free environment, typically resulting in greater amounts of inorganic nutrient forms and less organic materials (Kirchmann and Lundvall, 1993). Sutton et al. (1978) reported that anaerobically treated wastes generally contain a greater concentration of total Kjeldahl N (TKN) and NH4–N on a wet basis than aerobically treated wastes. In samples collected for a 2-yr period, J. Lorimor (2002, personal communication) found on average more of the TKN as NH4–N (measured as NH3–N) after swine manure went through anaerobic digestion for biogas production (76% NH4–N undigested manure and 89% NH4–N after digestion). Also, TKN was 10% greater with undigested manure than digested manure. Changes in total N and organic–inorganic forms with digestion could necessitate changes in agronomic utilization practices to correctly meet crop N needs and limit losses.
In many areas, swine production is being concentrated at both the regional and farm level (Bailey and Buckley, 2001). In the USA, the states of Iowa, North Carolina, Minnesota, and Illinois lead the country, accounting for >50% of total swine production. Also 52% of the 58 million swine population is fed in concentrated operations (5000 head or larger) (NASS, 2002). This controlled and large source of potential nutrients from these facilities must be utilized in some manner; most logically land application for crop nutrient use. Simultaneously, energy production before land application could help to reduce fossil fuel use and decrease the synthetic fertilizer budget. The objectives of this study were to evaluate swine manure that was anaerobically digested for biogas production as a source of N for corn production, compare N supply from manure before and after digestion, determine the effect of manure source and rate on soil inorganic N and corn plant N response, and determine manure N availability in the year of application and residual year.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Raw and anaerobically digested liquid swine manure was obtained from a commercial 5000 head sow-gestation-farrowing swine production facility located in southern Iowa. Manure in the facility is handled with shallow pull-plug gutters, and drained directly into a reception-transfer pit. From this reception-transfer pit manure is pumped into an anaerobic digester operated for biogas production. The anaerobic digested manure source was liquid swine manure that had been exposed to a 15-d residence digestion in the mesophilic anaerobic digester. The digested manure was collected as manure discharged from the digester. The raw manure was collected from the reception-transfer pit before entering the digester. For each application and manure source, the total volume needed was pumped into a transport tanker and hauled to the study site the day of application. The manure was off-loaded to a holding tank where it was thoroughly stirred before and during application. Samples were collected for analyses during transfer from the holding tank to the plot applicator. Analyses of two to four manure samples (Table 1) collected before each application were conducted by the Iowa State University Analytical Service Laboratory for NH3–N, total P, total K, TKN, total solids, volatile solids, chemical oxygen demand (COD), and pH (APHA, 1995). Nitrate-N was not included due to previous experience that indicated very little NO3–N in liquid swine manure.
|
15 cm, and resulted in a band of manure vertically and horizontally distributed. The applicator sweeps were pulled through the control plots (no manure application) to negate effects of injection tillage. No tillage was performed until spring before corn planting when the seedbed was prepared with tandem disking and field cultivation. The study utilized a split-split plot complete factorial treatment arrangement in a randomized complete block design, with four replications. Main plots (36.6 by 22.8 m) were manure sources (raw swine manure and anaerobically digested swine manure). Subplots (36.6 by 7.6 m) were manure N rate (none, low, high). The manure application volumes and total N rates are listed in Table 2. Sub-subplots (12.2 by 3.8 m) received surface broadcast NH4NO3 fertilizer at rates of 0, 45, 90, 135, 180, and 225 kg N ha–1 immediately after corn planting. The fertilizer N rates were in addition to the manure N applied (none, low, and high manure N rates, Table 2). No manure or fertilizer N was applied in the residual year.
|
Sampling and Analysis
Before manure application, all individual plots at both sites were sampled (0- to 15-cm depth) for routine soil test analysis. Average soil test results for the first site (manure applied in 2000 and 2002) were Bray and Kurtz P-1 60 mg kg–1, exchangeable K (1 M NH4OAc, pH 7.0) 174 mg kg–1, organic matter (LECO CHN-2000 analyzer; LECO Corp., St. Joseph, MI) 50 g kg–1, and pH (1:1 soil water paste) 6.0; and for the second site (manure applied in 2001) were Bray and Kurtz P-1 67 mg kg–1, exchangeable K 143 mg kg–1, organic matter 40 g kg–1, and pH 6.0 (soil test method procedures recommended for the North Central Region; Brown, 1998). The average P and K soil test values for the sites, as interpreted for corn production in Iowa, are classified as Very High for P and Optimum to High for K (Sawyer et al., 2002).
Soil samples were collected at a 0- to 15-cm depth from each manure source and rate sub-subplot treatment (no fertilizer N applied) in the center of the manure injection band (0 cm from the band) to monitor inorganic N within the manure band and midway between the bands (38 cm from the band) to provide background inorganic N concentrations. The center of manure bands were flagged after application. Two soil cores (1.9-cm diam.) were collected from two locations within each plot, with samples collected the first week of December 1999, second week of May and June 2000, and second week of May 2001. Because of frozen soils, samples could not be collected in December of 2000 and 2001. Also, after analysis of the first year samples, no differences between May and June samples were found. Therefore, in 2001 and 2002 samples were only collected in early spring. Field moist samples were sieved through a 5-mm sieve, placed in plastic bags, and frozen at –5°C until inorganic N (NH4–N and NO3 + NO2–N) analysis using accelerated diffusion (Khan et al., 1997). Moisture content was determined gravimetrically.
Soil samples were collected each year in early June to determine presidedress soil NO3–N concentrations (Blackmer et al., 1997) from selected treatments: 0, 90, and 180 kg N ha–1 fertilizer rates (no manure applied), and each manure source and rate (no fertilizer N applied). These treatments were chosen to evaluate the raw and digested manure rate effect and to compare with similar rates of fertilizer N. Samples were collected from the surface 0- to 30-cm soil layer in three sets of eight equally spaced cores (1.9-cm diam.) across the application spacing (Blackmer et al., 1997). The soil samples were dried in a forced-air oven at 50°C, ground to pass through a 2-mm sieve, and analyzed for NO3–N (Gelderman and Beegle, 1998).
Postharvest soil profile samples were collected for NO3–N determination from selected treatments: 0, 90, and 180 kg N ha–1 fertilizer rates (no manure applied), and each manure source and rate (no fertilizer N applied). Samples were not collected in the residual year. Two cores (5-cm diam.) were collected in 30-cm increments to a depth of 120 cm. Each sample was dried, ground to pass through a 2-mm sieve, and analyzed for NO3–N (Gelderman and Beegle, 1998).
A Minolta SPAD 502 chlorophyll meter (CM) (Konica Minolta, Ramsey, NJ) was used to evaluate the N status of corn plants during the growing season. All measurements were taken on 30 randomly selected plants from the middle two rows of each plot. Chlorophyll meter readings were taken at the V8, V15, R1, and R3 growth stages (Ritchie et al., 1993) using the procedure described by Peterson et al. (1993). Readings were taken from the newest fully expanded leaf with the collar exposed until the R1 stage, when the ear leaf was measured.
Aboveground biomass samples were collected at physiological maturity from the 0, 45, 90, 135, 180, and 225 kg N ha–1 rate sub-subplots where no manure was applied, and each manure source and rate (no fertilizer N applied). Samples were not collected in the residual year. Plants from 1.5 m in two rows (3 m total row length) were harvested from the sub-subplots, with plants cut just above soil line. The plants and ears were weighed, and then five plants (with ears removed) were chopped in the field. Subsamples of the chopped plant material and the five harvested ears were dried at 65°C to determine moisture content. The plant material and grain were ground to pass through a 1-mm screen, digested with H2SO4–H2O2 (Mills and Jones, 1996), and analyzed for total N by accelerated diffusion (Stevens et al., 2000). Total aboveground plant N was calculated from the plant and grain components (cob N not included).
Lower corn stalk samples were collected at physiological maturity from the same treatments and at the same time as the aboveground biomass samples. Ten stalk segments (15–35 cm above the ground) were collected per plot (Binford et al., 1990). The stalk samples were dried at 65°C and ground to pass through a 1-mm screen. Nitrate was extracted using the procedure described by Binford et al. (1992), with NO3–N determined by automated colorimetric flow-injection analysis (Lachat Instruments, Milwaukee, WI) (Mulvaney, 1996). Stalk samples were not collected in the residual year.
The middle rows of each plot were harvested for grain yield with a plot combine. Grain yields were adjusted to 155 g kg–1 moisture content. Grain samples were collected for protein determination by near-infrared spectroscopy (Rippke et al., 1995).
Analysis of variance was determined with PROC GLM (SAS Institute, 2001) for each application year and residual year using appropriate expected mean squares and F ratios for a three-factor treatment design in a randomized split-split plot design with main and interaction treatment effects considered fixed and blocks random. When the ANOVA indicated significance of fertilizer N rate (P 
0.10), for either manure source or the mean of manure sources, responses were regressed against fertilizer N rate. PROC NLIN or PROC GLM was used to investigate segmented quadratic- and linear-plateau regression models and continuous quadratic and linear models. If more than one model had a significant fit (P 0.10), the model with the largest R2 was chosen. The response fit was also visually inspected to confirm appropriate model choice. When manure source and rate treatments (with no fertilizer N) and all fertilizer N rates (with no manure) were sampled for plant N response, PROC GLM was used for separate analyses to determine significance (P 0.10) of manure source and rate, and fertilizer rate. Response to fertilizer N rate was fit with regression models as described earlier. When only selected treatments were sampled for soil inorganic N, PROC GLM was used to determine significance of treatment effects, and then Fisher protected least significant difference (FLSD, P 0.10) was calculated for means separation.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
Year of Application
Inorganic Nitrogen in Manure Application Zone
Elevated concentrations of NH4–N were found in soil samples collected from the center of the manure injection zone at the December 1999 sampling date, but not with spring sampling in any year (Table 3). It would be expected to initially find increased NH4–N due to the composition of the manure. Soil collected midway between the injection spacing should be outside the manure injection zone and represent nonmanure treated soil, and that NH4–N and NO3–N concentrations with that sample location would indicate background NH4–N and NO3–N at each sampling date (manure source and rate indicated by none in Table 3). However, with the relatively high concentrations measured at that location it is possible that manure with the sweep injection had some influence on inorganic N midway between the injection spacing.
|
Soil NO3
Each year soil NO3–N concentrations from early June samples (0- to 30-cm depth) were not different between raw and digested manure sources, however, manure rate and fertilizer N rate significantly affected NO3–N concentrations (Table 4). It is interesting that the concentrations from the manured plots are less than from the fertilized plots even though the applications supplied similar amounts of total N and that the manure sources had high NH4–N. Also, despite soil NO3–N concentrations being below the critical range of 16- to 25-mg NO3–N kg–1 for manured soils (Blackmer et al., 1997) each year with the high manure rate, adequate N was apparently available throughout the season for corn production as there was no yield response with additional fertilizer N (Table 5 and Fig. 2). A similar response occurred for the low manure rate in 2000. Others have noted less NO3–N present in the soil following manure injection than surface broadcast fertilizer treatments (Sutton et al., 1978; Randall et al., 1999). Suggested critical soil NO3–N concentrations in late spring are lower with manure application than fertilizer (Blackmer et al., 1997). Also, the manure was fall applied while the fertilizer treatments were spring applied. Differences in mineralization, immobilization, NO3–N movement out of the top 30-cm soil, or loss may explain part of the difference in soil NO3–N concentrations. The cooler and wetter spring in 2001 may have also resulted in differences between years.
|
|
|
|
Since manure source was considered not significant, to quantify manure rate effects on yield and N availability to corn, the manure source data were pooled. Quadratic-plateau regression equations (when significant) were fit to the N rate responses, and maximum plateau values were determined for the different manure rates (Fig. 2). Jokela (1992) suggested this method for quantifying the fertilizer N equivalence for a given rate of manure by developing best fit regression lines for the response variable (grain yield) to N fertilizer and determining the fertilizer rate that would produce similar yield as with manure. We have the benefit of having all fertilizer N rates applied in conjunction with each manure rate, which helps negate confounding from other manure effects on plant growth and yield.
For the no-manure regression fit lines, maximum yields (plateau join point) were obtained at 93 kg fertilizer N ha–1 in 2000, 116 kg fertilizer N ha–1 in 2001, and 128 kg fertilizer N ha–1 in 2002 (Fig. 2). In 2000, the low rate of manure (average 85 kg total manure N ha–1) resulted in a yield approximately the same as 90 kg N ha–1 from fertilizer, with no yield increase when fertilizer N was applied in addition to the manure N. This indicates equivalence of N supply from the fall-applied manure compared to the spring-applied fertilizer at 100% in 2000. High equivalence would be expected due to the large fraction as NH4–N in both manure sources, and conditions not conducive to N loss. In 2001, the maximum (plateau) yield occurred at 81 kg fertilizer N ha–1 with the regression fit line of the low manure rate (average 80 kg total manure N ha–1) and at 116 kg fertilizer N ha–1 with no manure. This suggests that the difference in N rates between the maximum yields in terms of N is the quantity delivered by the manure (35 kg N ha–1), which gives 44% equivalent N supply from the manure for the 2001 yr. In 2002, the maximum (plateau) yield occurred at 72 kg fertilizer N ha–1 with the regression fit line of the low manure rate (average 94 kg total manure N ha–1) and at 128 kg fertilizer N ha–1 with no manure. This suggests that the difference in N rates between the maximum yields in terms of N is the quantity delivered by the manure (56 kg N ha–1), which gives 60% equivalent N supply from the manure for the 2002 yr. The difference in equivalence of manure N supply could be attributed to varying growing seasons and loss potential from time of late fall application. Denitrification in the time period soon after application of liquid manure has been reported as an important mechanism for N loss (Comfort et al., 1990; Loro et al., 1997). There was no additional grain yield increase to fertilizer N when applied in conjunction with the high manure rate in any year.
The beneficial effect of manure N application on corn grain yield is reflected in the yield increase each year with manure application when no fertilizer N was applied (Fig. 2). The largest yield increase occurred with the low manure N rate, with smaller increase from the low to high manure N rate. Differences in yield increase from manure N application were found each year, which reflects differences in soil inorganic N supply, climate, and crop response to N.
Grain protein responded to manure and fertilizer N application each year. Manure rate and fertilizer N rate main effects, and the manure rate x N rate interaction, were significant each year (Table 5 and Fig. 2). Responses tended to follow those found for grain yield. The main difference was a slight increase in grain protein with fertilizer N application in addition to the high manure rate in 2001 and 2002, and less response to manure rate in 2001 and 2002. There was no differential response in grain protein between manure sources.
Chlorophyll Meter Readings
Leaf greenness, as determined by CM readings, is useful to monitor the in-season N status of corn because chlorophyll content is highly correlated with leaf N concentration (Schepers et al., 1992). As plant-available N increases, more leaf chlorophyll is produced and leaf greenness increases. Chlorophyll meter readings can be used to detect N deficiency, but leaf greenness and CM readings plateau at adequate to excess N, therefore limiting usefulness for determining excess N (Piekielek and Fox, 1992; Schepers et al., 1992; Wood et al., 1992).
Chlorophyll meter readings taken at V8, V15, VT and R3 growth stages were used to determine in-season corn N status and corn response to manure N sources and rates throughout the growing season. In all years the manure rate x N rate interaction (Table 5) was significant. No differences were found between raw and digested manure. Since no statistical differences between manure sources were found, means of the manure sources at different fertilizer N rates are shown in Fig. 3.
|
There was inconsistent and only minimal increase in CM readings with fertilizer N application in addition to the low manure rate in 2000, but there was consistent and greater CM reading response to fertilizer N with the low rate in 2001 and 2002. In those years, there was an increase in CM readings when fertilizer N was applied with the high manure rate, but the increase was minimal and readings with no fertilizer N were high. In all years, corn response to manure N was indicated by the increase in CM readings when no fertilizer N was applied (response similar to yield increase). Manure N supply, and associated magnitude of yield response, was not consistently predicted from the soil NO3–N test concentrations (0- to 30-cm depth) in early June, but was indicated better by response in CM readings (Table 4 and Fig. 3). With above-maximal fertilizer N application rates, CM readings tended to be greater with manure application. This may indicate a plant greenness response to something besides N, which could be other plant nutrients or unknown growth factors.
Aboveground Plant Nitrogen
Aboveground plant N uptake at maturity was not different between raw and digested manure (Table 5). Plant N was affected by manure rate in 2001 and 2002, but no statistically significant difference was found in 2000 (Table 5). In 2000, maximum total-N uptake was obtained at 88 kg fertilizer N ha–1, while no plateau was reached in 2001 and 2002 (Fig. 2). It is unknown why in 2001 and 2002 that the biomass N continued to increase after the point where maximum grain yields were obtained. Perhaps in 2001 and 2002, after the potential grain yield was set, plants continued to accumulate N in the biomass. This uptake did not influence grain yield, but grain protein did increase. These results are similar to CM readings at R3, and may represent luxury consumption as a result of a longer fall growing period with more available moisture. This is in contrast to rapid plant maturity in 2000 as a result of a few days of high temperatures and dry soil conditions. In 2000, the total-N uptake plateaued at a similar fertilizer N rate as grain yield. In 2001 and 2002, total-N uptake reflected the greater response to fertilizer N and low plant-available N with the low manure rate.
End of Season Corn Stalk NO3
Concentration of NO3–N in the lower stalk at maturity has the potential for evaluating excess N available to the corn plant (Binford et al., 1990). In 2000, stalk NO3–N was affected by manure rate and fertilizer N rate, while in 2001 and 2002 only by fertilizer N rate (Table 5 and Fig. 2). Manure source had no effect on stalk NO3–N concentration.
Across years, stalk NO3–N concentrations indicated excess available N from the high manure rate in 2000 and from the high fertilizer N rates in 2000 and 2001 (NO3–N concentration above 2000 mg kg–1) (Binford et al., 1992; Varvel et al., 1997; Brouder et al., 2000). In those years, the fertilizer N rate when stalk NO3–N indicated excess N was higher than the rate where yields reached maximum response (Fig. 2). In 2002, stalk NO3–N increased slightly with high fertilizer N rates, but was well below 2000 mg kg–1, even with N application in excess of the maximal response rate. Stalk NO3–N concentrations were increased significantly with manure application in 2000, but not in 2001 and 2002. This indicates less manure N supply in 2001 and 2002 and possibly the effect of late-season N accumulation into grain.
Residual Year
Soil NO3
In the year after manure and fertilizer application (2001 and 2002), soil NO3–N concentrations in the top 30-cm of soil were well below the 16- to 25-mg NO3–N kg–1 critical concentration range (Blackmer et al., 1997) with all prior-year manure and fertilizer rates (Table 7). In 2001, concentrations were significantly increased where manure had been applied the prior year, but the increases were small (1–3 mg kg–1). In 2002, differences were small and variable, with the largest concentration where the 180 kg N ha–1 fertilizer rate had been applied the prior year. Overall, soil NO3–N concentrations indicated little NO3 production in the spring of the residual year or carryover from prior year applications. Differences between prior year manure rates were minimal. This indicates that the soil NO3–N concentration in the top 30 cm of soil in early June is not a sensitive indicator of residual N availability, or of excess manure or fertilizer N applied to corn in the prior year. The large profile NO3–N measured in the fall of 2001 with the 180 kg N ha–1 fertilizer rate (Table 6) was indicated in the soil NO3–N concentration the following year (June 2002; Table 7), but the increase in the June 2002 soil NO3–N concentration compared with no applied N was only 3 mg kg–1.
|
|
|
Chlorophyll Meter Readings
Chlorophyll meter readings collected throughout the growing season indicated increased N supply with increasing prior year manure and fertilizer application rates (Tables 8 and 9). This indicates that an enhanced inorganic N supply was available early in the growing season. There was no significant difference between manure sources. In 2002, the manure rate effect on CM readings in the residual year was not significant after the V8 growth stage. This indicates that the supply of N from the prior year manure application was from carryover NO3 only, and not a residual release of inorganic N from manure organic material. The increase in CM readings had the same trend as yield and grain protein response to the prior year manure and fertilizer applications.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. W. Singer, C. A. Cambardella, and T. B. Moorman Enhancing Nutrient Cycling by Coupling Cover Crops with Manure Injection Agron. J., November 7, 2008; 100(6): 1735 - 1739. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. H. Chantigny, D. A. Angers, G. Belanger, P. Rochette, N. Eriksen-Hamel, S. Bittman, K. Buckley, D. Masse, and M.-O. Gasser Yield and Nutrient Export of Grain Corn Fertilized with Raw and Treated Liquid Swine Manure Agron. J., August 11, 2008; 100(5): 1303 - 1309. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Crop Science | Vadose Zone Journal | |||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
, *, **, and *** for 0.10, 0.05, 0.01, and 0.001 probability levels, respectively.