SSSAJ Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 21 June 2006
Published in Soil Sci Soc Am J 70:1359-1369 (2006)
DOI: 10.2136/sssaj2005.0298
© 2006 Soil Science Society of America
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 Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Golden, B. R.
Right arrow Articles by DeLong, R. E.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Golden, B. R.
Right arrow Articles by DeLong, R. E.
Agricola
Right arrow Articles by Golden, B. R.
Right arrow Articles by DeLong, R. E.
Related Collections
Right arrow Nitrogen

Nutrient Management & Soil & Plant Analysis

Recovery of Nitrogen in Fresh and Pelletized Poultry Litter by Rice

Bobby R. Goldena,*, Nathan A. Slatona, Richard J. Normanb, Edward E. Gbur, Jr.c, Kristofor R. Bryeb and Russell E. DeLonga

a Dep. of Crop, Soil, and Environmental Sciences, Univ. of Arkansas, 1366 West Altheimer Drive, Fayetteville, AR 72704
b Dep. of Crop, Soil, and Environmental Sciences, Univ. of Arkansas, 115 Plt. Sci. Bldg., Fayetteville, AR 72701
c Agric. Statistics Lab., Univ. of Arkansas, Fayetteville, AR 72701

* Corresponding author (bgolden{at}uark.edu)


   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 SUMMARY
 REFERENCES
 
The N-fertilizer value of preplant incorporated poultry litter for rice (Oryza sativa L.) grown using the direct-seeded, delayed-flood (DF) production system is unknown. The research objective was to compare preplant incorporated fresh (FPL) and pelleted (PPL) poultry litter with urea applied preflood to determine the urea-nitrogen (N) equivalence of poultry litter for rice. Fresh and pelleted litter were preplant incorporated at total-N rates ranging from 34 to 270 kg N ha–1 and compared with urea applied preflood at rates ranging from 34 to 168 kg N ha–1 at five site-years on silt loam soils. Net-N uptakes and grain yields between FPL and PPL were similar, but significantly lower when compared with urea applied preflood, regardless of site-year. By heading, when averaged across all N rates and site-years, rice recovery of the urea-N applied preflood averaged 76%. In contrast, the apparent recovery of N applied as FPL or PPL averaged only 14%. Net-grain yields for rice fertilized with urea increased nonlinearly as N rate increased with near maximal yields produced with ≥101 kg urea-N ha–1. Grain yields for FPL and PPL increased linearly and approached the near maximal yields produced with urea-N only when 270 kg total-N ha–1 was applied. Based on net grain yield, multiplying the total-N content in FPL and PPL by 0.25 reasonably estimates its equivalence to urea applied preflood. Recommended preflood urea-N rates should be decreased by 0.25 times the total N applied as poultry litter.

Abbreviations: DF, delayed flood • FPL, fresh poultry litter • HDG, heading • NF, normal flood • PD, panicle differentiation • PPL, pelleted poultry litter • PTBS, Pine Tree Branch Station • PAN, plant-available nitrogen • RREC, Rice Research Extension Center


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 SUMMARY
 REFERENCES
 
POULTRY LITTER has traditionally been applied to satisfy the N requirements of forages in fields near poultry houses in western Arkansas. Nitrogen-based application rates of poultry litter have caused soil-test phosphorus (P) to accumulate to high levels in many soils used for pasture and hay forage production. High soil-test P and the P in poultry litter have been implicated as primary contributors to reduced water quality in northwest Arkansas (Pote et al., 1996). Nutrient management planning and litigation have imposed restrictions on the fields that can receive fertilizers or soil amendments that contain P and the amount of P that can be applied. The area that may receive P applications in western Arkansas is not sufficient for the amount of poultry litter and other animal manures produced annually (Slaton et al., 2004).

Sims (1997) described several alternative uses for excess poultry litter with one option being to transport and apply the litter to soils used for row-crop production. In Arkansas, the areas of intensive poultry and row crop production are separated geographically in the western and eastern halves of the state, respectively (Slaton et al., 2004). The feasibility of transporting litter is partially dependent on its value as a fertilizer source for row crops (Bosch and Napit, 1992).

In eastern Arkansas, rice is grown on about 614 000 ha annually (AASS, 2003) with 99% of the rice area receiving 162 kg N ha–1 and 44% receiving 27 kg P ha–1 (USDA-NASS, 2001). Soils used for rice production typically have low soil-test P (DeLong et al., 2004), are also used for irrigated-soybean [Glycine max (Merr.) L.] production, and are frequently precision-graded to improve irrigation efficiency (Cooke et al., 1996). The primary use of FPL in eastern Arkansas has been as a soil amendment to aid in restoring productivity to recently precision-leveled soils (Miller et al., 1990), which represents about 91 Mg litter yr–1 (Kellogg et al., 2000). Also, PPL is currently being transported to eastern Arkansas from Delaware and applied at low to moderate rates (400 – 2000 kg ha–1) to undisturbed and leveled soils used for rice production. The University of Arkansas currently has no recommendation that accounts for the plant-available N (PAN) in poultry litter for flood-irrigated rice. Knowledge of the N-fertilizer value of FPL and PPL applied to soils used for flood-irrigated rice production may facilitate its use in eastern Arkansas as a fertilizer source, allow for the development of inorganic-N rate adjustments when poultry litter is applied as a soil amendment, and facilitate its transport from western Arkansas.

The PAN content of poultry litter has been described for upland crops such as corn (Zea mays L.). Bitzer and Sims (1988) reported that the amount of organic N mineralized from 20 FPL sources averaged 670 g kg–1 (N mineralized based on total organic N) during a 140-d aerobic incubation study. Sims (1987) and Bitzer and Sims (1988) showed FPL applied at PAN rates calculated as 100% of the inorganic N and 60% of the organic N contents generally produced similar corn grain yields as like rates of NH4NO3–N. Cooperband et al. (2002) showed similar results with corn by calculating PAN as 80% of the inorganic N and 50% of the organic N. Row-crop responses to PPL have not been reported, but incubation studies conducted by Hadas et al. (1983) and Cabrera et al. (1993) suggest that PPL and FPL mineralize similar amounts of organic N. Literature suggests that poultry litter has appreciable value as a N-fertilizer source for upland crops. However, few studies comparing N uptake and yield of flood-irrigated rice grown using recommended N-fertilization practices with rice receiving poultry litter or other animal manures as an N source have been reported.

Singh et al. (1997) showed that transplanted rice receiving poultry manure 4 d before transplanting produced lower grain yields than rice receiving like rates of urea-N during the first year of a 3-yr study. Based on rice grain yield data, they estimated that the urea-equivalent N rate of poultry manure could be calculated by multiplying the total-N rate of poultry manure by 0.58 to 0.88 making its PAN value similar to upland crops. However, Singh et al. (1997) also reported that rice recovery of urea-N was only about 44% of the applied N, which is less efficient than urea-N recovery by rice in Arkansas.

Rice grown in the direct-seeded, delayed-flood production system used in Arkansas commonly recovers 65 to 72% of the urea-N applied preflood to a dry soil surface and followed by application of the permanent flood (Norman et al., 2003). The N-fertilizer value of poultry litter used to produce transplanted rice as determined by Singh et al. (1997) may not be representative for the rice-production practices used in Arkansas. The direct-seeded, DF production system consists of drill-seeding rice followed by establishing a 10-cm deep flood at the 5-leaf stage, which is about 5 to 8 wk after seeding. The flood is maintained for the duration of the growing season and removed about 2 wk before the anticipated harvest date. The rapid mineralization of organic N in FPL and PPL may produce ample amounts of PAN or inorganic N under moist, upland conditions (Bitzer and Sims, 1988). However, a proportion of the mineralized N will undergo nitrification and be present as NO3–N before rice fields are flooded. Upon flooding, a large proportion of the NO3–N will subsequently be lost via denitrification (Patrick and Wyatt, 1964) which will reduce the N-fertilizer value of poultry litter for flood-irrigated rice.

The primary research objective was to determine the urea N-fertilizer value of preplant incorporated poultry litter compared with the standard practice of applying urea post-emergence to a dry soil surface followed by flooding. A second objective was to determine whether poultry litter form, fresh or pelletized litter, has a significant influence on the N-fertilizer value of poultry litter. The ultimate research goal was to develop inorganic N-rate adjustment recommendations that account for the PAN from poultry litter recovered by rice. We hypothesized that (i) rice receiving fresh or pelletized litter would have similar N uptakes and yields that were significantly lower compared with urea applied preflood, and (ii) the urea-N equivalence of FPL and PPL, as measured by N uptake and grain yield, would be less than the PAN values of 50 to 60% which are commonly used for upland crops.


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 SUMMARY
 REFERENCES
 
Description of Sites
Five field experiments, two in 2003 and three in 2004, were conducted to determine the N-fertilizer value of FPL and PPL for flood-irrigated rice. Experiments at the Rice Research and Extension Center (RREC, N 34° 27.882' lat; W 091° 25.063' long) were established during 2003 (03) and 2004 (04) on a Dewitt silt loam (fine, smectitic, thermic Typic Albaqualfs). At the Pine Tree Branch Station (PTBS, N 35° 07.269' lat; W 090° 57.428' long) studies were conducted in 2003 and 2004 on a Calhoun silt loam (fine-silty, mixed, active, thermic Typic Glossaqualfs). A single experiment was established in 2003, but, in 2004, two adjacent studies that differed only in floodwater management were established at the PTBS. The permanent flood was established at the normal time (NF, 5-leaf stage) or delayed (DF, 5-leaf + 9 d). The flood was applied at the normal time for both studies conducted during 2003. All studies will be identified by the abbreviations corresponding to the site, year, and flood management system (Table 1).


View this table:
[in this window]
[in a new window]
  		Table 1. Selected soil chemical property means (n = 8) of research sites managed with a normal (NF) or delayed flood (DF) at the Pine Tree Branch Station (PTBS) and Rice Research Extension Center (RREC) in 2003 (03) and 2004 (04).

 
Soybean was the previous crop grown at all site-years. Eight composite soil samples (two per replicate) were collected from the 0- to 10-cm depth at each site before application of poultry litter. Each composite sample consisted of eight, 2.5-cm diameter cores. Soil samples were oven dried, crushed to pass through a 2-mm sieve, and extracted with Mehlich-3 (Mehlich, 1984). Mehlich-3 extracts were analyzed using inductively coupled atomic plasma spectroscopy (Soltanpour et al., 1996). Soil water pH was determined in a 1:2 soil weight/water volume ratio using a glass electrode. Total N and C were determined by combustion (LECO CN2000, St. Joseph, MI; Nelson and Sommers, 1996). The mean values of selected soil chemical properties are listed in Table 1.

Characterization of Poultry Litter
Fresh poultry litter was obtained from the University of Arkansas Savoy poultry production facility in 2003 and 2004. Poultry litter was removed directly from the poultry house following 18 and 12 mo of broiler production in March 2003 and 2004, respectively. The bedding material consisted of nearly equal proportions of rice hulls and saw dust. Pelletized poultry litter was obtained from Plant Right Inc. (Purdy, MO) in 2003 and from Lee Harris Farms and Company (Bentonville, AR) in 2004. (Note: Plant Right Inc. is no longer in business.) After collection, FPL and PPL were stored in sealed 67.5 L Rubbermaid containers. From each litter source, three composite samples were analyzed for nutrient content on an ‘as is’ or moist basis (Table 2). Samples were homogenized and particle size was reduced in preparation for chemical analyses using a coffee bean grinder (Model CBG5, Black and Decker, Towson, MD). Total N and C were determined by combustion (LECO CN2000, St. Joseph, MI) as described by Watson et al. (2003). Litter NH4–N and NO3–N concentrations were determined by extracting a 0.5-g subsample of ground litter with 2 M KCl and the NH4–N and NO3–N concentrations of filtrates were determined by colorimetery using an autoanlayzer (SKALAR, San + Segmented Flow, Norcross, GA; Peters et al., 2003). The concentrations of P, K, and other elements were determined by digestion of a 0.5- to 1.0-g subsample of ground litter using the concentrated HNO3 and 30% (w/w) H2O2 method as described by Kovar (2003). The elemental concentrations of digests were determined by ICPS. After the analyses were completed, FPL and PPL were weighed into large Ziploc (SC Johnson, Racine, WI) freezer bags which were sealed and stored at room temperature until the treatments were applied.


View this table:
[in this window]
[in a new window]
  		Table 2. Selected chemical properties of fresh (FPL) and pelleted (PPL) poultry litter used in field studies conducted during 2003 (03) and 2004 (04).

 
Treatments
Individual plots, measuring 2.0 m wide x 4.9 m long were flagged to establish plot boundaries. Fresh poultry litter was broadcast with a lawn spreader (Scott's Accugreen 3000, Stevensville, MI) and PPL was broadcast by hand to randomly assigned plots at total-N rates of 34, 68, 134, 202, and 270 kg N ha–1. Phosphorus (20 kg P ha–1 as triple superphosphate) and K (50 kg K ha–1 as muriate of potash) fertilizers were broadcast to all plots that were to receive no FPL or PPL. Zinc (11 kg Zn ha–1 as ZnSO4) was also applied to experiments at the PTBS because of high soil pH. Within 5 h after application of FPL and PPL all plots were mechanically tilled to a depth of 4 cm with a rotary tiller at all site-years, except PTBS03-NF where tillage was performed with a field cultivator. Nitrogen rates for FPL and PPL were based on estimates that 45 to 75% of the organic N would mineralize into PAN during the year of application (Castellanos and Pratt, 1981; Hadas et al., 1983; Bitzer and Sims, 1988; Gordillo and Cabrera, 1997). Greater N rates from FPL and PPL were not considered because the 270 kg N ha–1 rate also supplied 83 to 104 kg P ha–1, which would not be recommended for rice. The range of litter-N rates was sufficient to provide linear or nonlinear response curves for N uptake and grain yield which could be compared with inorganic-N fertilizer.

The long-grain rice cultivar Wells was drill-seeded into conventionally tilled seedbeds at 112 kg seed ha–1 within 1 d after poultry litter incorporation. Each plot contained nine rows of rice spaced 19 cm apart and was surrounded by a 45-cm wide alley that contained no rice. Near the 5-leaf stage, urea (460 g N kg–1) was broadcast to a dry soil surface to plots that had received no poultry litter. Urea was applied at rates of 34, 67, 101, 134, and 168 kg N ha–1 for all site-years. Norman et al. (1999) reported that Wells rice produced maximal grain yields when urea was applied at the 5-leaf stage (i.e., preflood) at rates ranging from 100 to 135 kg N ha–1 for silt-loam soils.

Following urea application, a 10-cm deep permanent flood was established and maintained until rice reached physiological maturity. For the PTBS04-DF site, urea application and establishment of the permanent flood were delayed 9 d to provide some variation in irrigation water management. The dates of several agronomic events are listed in Table 3. Each year, weeds were controlled by use of a tank mixture of 0.30 kg ha–1 clomazone [2-(2-chlorobenzyl)-4,4-dimethylisoxazolidin-3-one] plus 0.40 kg ha–1 quinclorac (3,7-dichloroquinoline-8-carboxylic acid) applied to the soil surface before rice emergence, followed by a tank mixture of 4 kg ha–1 propanil (3,4-dichloropropananilide) plus 0.04 kg ha–1 bensulfuronmethyl {methyl2-[(4,6-dimethoxypyrimidin-2-yl) carbamoyl-sulfamoyl-methyl] benzoate} at the 4-leaf stage. In general, rice management closely followed the University of Arkansas Cooperative Extension Service recommendations for stand establishment, pest management, and irrigation management (Slaton, 2001).


View this table:
[in this window]
[in a new window]
  		Table 3. Selected dates of agronomic management events for research trials managed with a normal (NF) or delayed flood (DF) at the Pine Tree Branch Station (PTBS) and Rice Research Extension Center (RREC) in 2003 (03) and 2004 (04).

 
Measurements
Aboveground plant samples were taken at panicle differentiation (PD, 1.2 cm internode elongation) and late-boot to early-heading (HDG, Table 3). Panicle differentiation occurs about 4 wk after the 5-leaf stage and flood establishment, and represents maximal uptake of urea-N applied preflood by rice grown in the DF production system (Wilson et al., 1989). The HDG stage represents maximal N accumulation by rice during the growing season (Guindo et al., 1994b). Plant samples were taken from a 0.9-m section in the first inside row of each plot. For the PTBS04-DF site-year, plant samples were taken only at HDG. Samples were dried at 60°C in a forced-draft oven, weighed, and ground to pass through a 1-mm sieve. A 0.10- to 0.15-g subsample was weighed into a tared 7-mm Alphagel gel capsule (Alpha Resources, model AR 338, Stevensville, MI) and total N was determined by combustion (LECO FP-248, St. Joseph, MI; Campbell, 1992).

Aboveground N content was calculated by multiplying the whole-plant N concentration with total aboveground dry matter accumulation. The net-N uptake by rice was calculated using the difference method. For each growth stage, the N uptake of the unfertilized control (0 kg N ha–1) was subtracted from total-N uptake of treatments receiving urea, FPL, and PPL. Nitrogen uptake data represents net-N uptake above the unfertilized control attributed to each N source and rate combination.

Grain yield was determined at physiological maturity by harvesting a 3- to 5-m2 area from the middle of each plot with a small plot combine. Grain weights and moisture contents were recorded and grain yields were adjusted to a uniform moisture content of 120 g kg–1 for statistical analysis. Net grain yield above the unfertilized control was calculated as described previously for net-N uptake.

Statistical Analysis
Each experiment was a randomized complete block with treatments defined by three N sources with each source applied at five N rates. The N rates for PPL and FPL were identical, but urea-N was applied at different and lower total-N rates. An unfertilized control (0 kg ha–1) was included in each replication. Each treatment was replicated four times. Treatment means for net-N uptake at PD and HDG and net grain yield were calculated across replicates for each site-year.

There were five site-years for net-N uptake at HDG and grain yield, but only four site-years for net-N uptake at PD. The net-N uptake and grain yields were initially regressed on N-rate allowing for both linear and quadratic terms with coefficients depending on N-source and site-year. Nonsignificant (P > 0.05) model terms were removed sequentially and the model was refit until a satisfactory model was obtained. Differences among all remaining coefficients, which varied by N source, site-year, or both were determined using single degree of freedom contrasts. All statistical analyses were conducted using SAS version 9.1 (SAS Institute Inc., Cary NC).


   RESULTS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 SUMMARY
 REFERENCES
 
Visual Observations
Poultry litter visually increased rice seedling and weed growth between emergence and the 5-leaf stage at all site-years compared with rice that was to receive urea preflood. Rice fertilized with FPL and PPL were visually taller and greener before flooding than rice that received no poultry litter. At PTBS04-NF and PTBS04-DF, plots receiving poultry litter contained more and larger pigweeds (Amaranthus palmer L.) than plots receiving no FPL and PPL suggesting that FPL and PPL stimulated pigweed seed germination and early-season growth. If weed seed germination and growth are enhanced by preplant poultry litter applications, weed control may be more difficult; the timeliness of weed control may be more important in preventing crop yield losses from weed competition and may warrant further investigation. Weeds were effectively controlled before flooding with the herbicide program used at all site-years.

By 2 wk after flooding, rice receiving preflood urea was green and growing vigorously. In contrast, plants receiving all but the highest N rate of FPL and PPL applied preplant were pale green indicating N deficiency. By 3 wk after flooding, rice receiving the highest FPL and PPL N rate and the two lowest urea-N rates also showed signs of N deficiency. In general, visual observations indicated that post-flood rice growth increased as N rate increased within each N source, and that the growth of rice receiving preflood urea was greater than rice receiving FPL and PPL applied preplant-incorporated, which had equal growth.

Net Nitrogen Uptake at Panicle Differentiation
Net-N uptake by rice was measured at the PD stage which occurred 68 to 71 d after poultry litter application and 21 to 26 d after the permanent flood was established following urea application (Table 3). Nitrogen uptake at PD by rice receiving no N as urea, FPL, or PPL averaged 23 kg N ha–1 for PTBS03-NF, 24 kg N ha–1 for PTBS04-NF, 16 kg N ha–1 for RREC03-NF, and 17 kg N ha–1 for RREC04-NF.

Net aboveground N uptake increased linearly as N rate increased for each N source (Table 4 and Fig. 1). The slope coefficients for net-N uptake varied among N sources, but were uniform across site-years within each N source (Table 5). Net-N uptake at PD was also significantly affected by site-year, an intercept term. The slope for urea was significantly greater than the slopes for FPL and PPL, which were numerically different but statistically similar. Net-N uptake from the greatest application rates of FPL and PPL ranged from 11 to 39 kg N ha–1 compared with 99 to 122 kg N ha–1 for 168 kg urea-N ha–1 (Fig. 1). The lowest urea-N rate, 34 kg N ha–1, produced numerically similar rice plant-N contents as FPL and PPL rates ranging from 134 to 270 kg N ha–1. The linear slope coefficients indicate that, when averaged across all N rates within each source, rice recovered 68% of the urea-N applied and 7 to 9% of the total N applied as FPL and PPL.


View this table:
[in this window]
[in a new window]
  		Table 4. Analysis of variance p-values for net-N uptake at panicle differentiation (PD) and heading (HDG) and net rice grain yield as affected by N source (NS), N rate (NR), site-year (SY), and their significant interactions for five experiments conducted on silt loam soils during 2003 and 2004.

 

Figure 1
View larger version (22K):
[in this window]
[in a new window]
  		Fig. 1. Net-N uptake by rice at panicle differentiation as affected by N rate and N source for research trials managed with a normal (NF) flood at the Pine Tree Branch Station (PTBS) and Rice Research Extension Center (RREC) in 2003 (03) and 2004 (04) on silt loams.

 

View this table:
[in this window]
[in a new window]
  		Table 5. Regression coefficients for net-N uptake (kg N ha–1) at panicle differentiation as affected by N rate for research trials managed with a normal (NF) or delayed flood (DF) at the Pine Tree Branch Station (PTBS) and Rice Research Extension Center (RREC) in 2003 (03) and 2004 (04).

 
Net Nitrogen Uptake at Heading
Net-N uptake was measured at HDG, which occurred 23 to 28 d after samples were taken at PD on silt-loam soils, to estimate maximal total N uptake (Table 3). Nitrogen uptake at HDG by rice receiving no N as urea, FPL, or PPL averaged 47 kg N ha–1 for PTBS03-NF, 56 kg N ha–1 for PTBS04-NF, 66 kg N ha–1 for PTBS04-DF, 36 kg N ha–1 for RREC03-NF, and 45 kg N ha–1 for RREC04-NF.

The general trends for net-N uptake among N sources at HDG were comparable to those described for the PD stage (Table 4). Within each N source, net-N uptake generally increased linearly as N rate increased, but the slope coefficient varied among site-years (Table 6 and Fig. 2). Within each site-year, the slope for urea was always significantly greater than the slopes for FPL and PPL, which were not different. The magnitude of the slope coefficients (i.e., high to low) of the three N sources followed the same pattern across site-years. For example, the numerical slope coefficients for urea, FPL, and PPL were highest at RREC04-NF and lowest at RREC03-NF (Table 6). All slopes except for FPL and PPL at RREC03-NF, FPL at PTBS03-NF, and FPL at PTBS04-DF were significantly greater than zero. Across the five site-years, net-N uptake at HDG (Fig. 2) showed a greater range than N uptake at PD (Fig. 1). Net-N uptake of urea-N among site-years ranged from about 80 to 140 kg N ha–1 for 168 kg urea-N ha–1 compared with 24 to 84 kg N ha–1 for 270 kg total-N ha–1 applied as FPL or PPL (Fig. 2). The range of net-N uptakes from the highest rate of FPL and PPL were numerically comparable to the range of net N uptakes from urea rates of 34 to 101 kg N ha–1. The linear slope coefficients indicate that, when averaged across all N rates within each source and each site-year, rice recovered 69 to 87% (average = 76%) of the urea-N applied and <25% (average = 14%) of the total N applied as FPL and PPL.


View this table:
[in this window]
[in a new window]
  		Table 6. Regression coefficients for net-N uptake at early heading as affected by N rate for research trials managed with a normal (NF) or delayed flood (DF) at the Pine Tree Branch Station (PTBS) and Rice Research Extension Center (RREC) in 2003 (03) and 2004 (04).

 

Figure 2
View larger version (24K):
[in this window]
[in a new window]
  		Fig. 2. Net-N uptake by rice at heading as affected by N rate, N source, and site-year for research trials managed with a normal (NF) or delayed flood (DF) at the Pine Tree Branch Station (PTBS) and Rice Research Extension Center (RREC) in 2003 (03) and 2004 (04) on silt loams.

 
Net Rice Grain Yield
The mean grain yields of rice receiving no urea, FPL, or PPL averaged 5356 kg ha–1 for PTBS03-NF, 5496 kg ha–1 for PTBS04-NF, 6078 kg ha–1 for PTBS04-DF, 3943 kg ha–1 for RREC03-NF, and 4759 kg ha–1 for RREC04-NF. Net grain yield of rice was a nonlinear (quadratic) function of N rate which differed by N source and only the intercepts depended on site-year (Table 4). Net grain yields for rice fertilized with urea increased in a significant nonlinear fashion as N rate increased (Table 7 and Fig. 3). Net grain yields increased as the urea-N rate increased from 34 to 101 kg urea-N ha–1 then reached a plateau from 101 to 168 kg urea-N ha–1. For FPL and PPL, the nonlinear slope coefficients did not differ from zero indicating that grain yields increased linearly across the range of applied total-N rates. The nonlinear and linear coefficients for urea were significantly greater than the coefficients for FPL and PPL.


View this table:
[in this window]
[in a new window]
  		Table 7. Regression coefficients for net rice grain yield as affected by N rate for research trials managed with a normal (NF) or delayed flood (DF) at the Pine Tree Branch Station (PTBS) and Rice Research Extension Center (RREC) in 2003 (03) and 2004 (04).

 

Figure 3
View larger version (25K):
[in this window]
[in a new window]
  		Fig. 3. Net rice grain yield at physiological maturity as affected by N rate and N source for research trials managed with a normal (NF) or delayed flood (DF) at the Pine Tree Branch Station (PTBS) and Rice Research Extension Center (RREC) in 2003 (03) and 2004 (04) on silt loams.

 
The different trends describing net grain yield between poultry litter sources applied preplant and urea applied preflood indicate that each unit (i.e., kg) of total-N in poultry litter is a less efficient N source compared with urea. The predicted preflood urea-N rate required to produce maximum agronomic rice yields was 139 kg urea-N ha–1. The near maximal mean net yields produced with 101 kg urea-N ha–1 ranged from 3109 to 6132 kg ha–1 among site-years. In contrast, application of 270 kg total-N ha–1 as FPL or PPL produced net grain yield increases ranging from 2102 to 5812 kg ha–1. The net yields for FPL and PPL did not reach a plateau in these studies, which indicates that ≥270 kg total-N ha–1 of FPL and PPL (about 6500 kg moist basis FPL or PPL ha–1) would be needed to produce maximal rice grain yields on most silt loams.


   DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
SUMMARY
 REFERENCES
 
Poultry litter is a suitable N source for corn (Sims, 1987; Bitzer and Sims, 1988, Cooperband et al., 2002) and cotton (Sistani et al., 2004; Mitchell and Tu, 2005) when applied at the appropriate rates, providing an alternative to inorganic-N fertilizer in areas where row crops are produced in close proximity with poultry. However, five site-years of data suggest that FPL and PPL alone are not adequate N-source alternatives to urea applied preflood for flood-irrigated rice grown with the direct-seeded, DF system. Differences in management, especially with regards to irrigation and rainfall, of rice and most upland crops influence the time and processes that occur in the N cycle which influence the ultimate availability of N. Although leaching, runoff, volatilization, and denitrification are common and potential N-loss pathways for cultivated cropland that requires N fertilization, the potential for denitrification is always present in rice grown using flood irrigation when NO3–N is present.

Mineralization of organic N from poultry litter occurs quite rapidly once mixed with soil (Castellanos and Pratt, 1981; Hadas et al., 1983; Bitzer and Sims, 1988; Gordillo and Cabrera, 1997). The time between litter application and establishment of the permanent flood in these studies ranged from 42 to 59 d (Table 3) and represents sufficient time for appreciable mineralization of organic N, as well as nitrification of mineralized organic N (Sims, 1986). Differences in net-N uptake among site-years were expected because mineralization of poultry litter (Sims, 1986) and organic matter (Brye et al., 2003), as well as the rate of nitrification (Chapman and Liebig, 1952; Stanford and Smith, 1972) may vary among soils, environmental conditions (Hadas et al., 1983), or both.

Organic N from FPL and PPL that had been mineralized and undergone nitrification was likely lost before establishing the permanent flood via denitrification following significant rainfall events or when rice was flush irrigated or after the permanent flood was established. Soil NO3–N is usually depleted within 10 to 14 d after flooding (Patrick and Wyatt, 1964; Norman et al., 1989). Assuming that little or no denitrification occurred before flooding in these studies, rapid N loss from denitrification after flooding would prohibit efficient use of the mineralized organic N that had undergone nitrification because rapid uptake of N by rice does not begin for at least 7 d after flooding (Guindo et al., 1994b).

Potentially available N can be defined as the net N mineralized from poultry litter following its application and is usually categorized by cropping season. Bitzer and Sims (1988) estimated that 80% of the inorganic N and 60% of the organic N in FPL would be plant available during the first year, based on incubation studies. Field studies with corn confirmed that their estimate of PAN was appropriate for corn grown in Delaware. Bitzer and Sims (1988) estimate of PAN in FPL is reasonably accurate for most temperate growing environments. However, estimates of PAN do not always accurately reflect the efficiency of mineralized N uptake by the crop to be grown. Alternatively, comparisons of net-N uptake or grain yield among N sources may account for crop management factors, such as application time, irrigation method, or soil texture that may influence N recovery, and can be expressed as N-fertilizer equivalents, which for rice would be preflood urea-N equivalents.

Uptake of N at PD and HDG and grain yields of rice receiving preplant incorporated FPL and PPL were lower than for urea applied preflood at all site-years (Fig. 1Go–3). The majority of net-N uptake from FPL and PPL at PD likely represents mineralized organic N that had not undergone nitrification and was absorbed by rice shortly after flooding. Compared with urea, net-N uptakes and grain yields were lower for FPL and PPL despite greater total-N application rates (Fig. 1Go–3). Uptake of preplant incorporated urea in the direct-seeded, DF rice culture is also low. Norman et al. (1989) reported rice uptake of preplant incorporated urea–15N was only 16 to 23% at maturity compared with 51 to 61% of the urea–15N applied in three split applications following recommended practices. Use of a nitrification inhibitor with preplant incorporated urea–15N increased N recovery to 34 to 44%, suggesting that much of the preplant incorporated N was lost via denitrification before or shortly after flooding.

Rice recovery of urea applied preflood to silt-loam soils at the PD stage usually ranges from 65 to 72% when the urea is applied to a dry soil and flooded immediately (Norman et al., 2003), and is achieved within 3 to 4 wk after flooding (Wilson et al., 1989; Guindo et al., 1994a). At PD, the recovery of urea-N averaged 68% for the five silt-loam soils in our study. In comparison, recovery of the total N applied as FPL or PPL averaged 8%. This suggests that, by PD, urea-N is utilized 8.5 times more efficiently than the total N in poultry litter. Sims (1987) reported that N uptake by corn was greater when fertilized with NH4NO3 compared with corn receiving poultry manure in 1 of 3 yr. The low uptake of N from FPL and PPL by rice during vegetative growth would be problematic for its use as an early-season N source because sufficient N uptake by PD is critical for the production of high-yielding rice (Bollich et al., 1994; Wilson et al., 1998; Norman et al., 1999).

By HDG, the stage of maximal N uptake by rice (Guindo et al., 1994b), the apparent recovery of urea-N applied preflood ranged from 69 to 87% (average = 76%) among site-years, which is high but typical N recovery for urea applied preflood and managed properly. The average apparent recovery of the total N applied as FPL and PPL was <25% across the five site–years. Net-N uptake data suggest that 5 kg total-N ha–1 applied preplant as poultry litter is roughly equal to 1 kg urea-N ha–1 applied preflood. Therefore, only about 18% of the total N in preplant incorporated poultry litter is equivalent to urea applied preflood. Total-N uptake by corn receiving inorganic-N fertilizer is usually greater than similar total-N rates applied as poultry litter (Sims, 1987; Cooperband et al., 2002). Estimates of PAN are used to account for the relative availability of manure N in comparison to inorganic-N fertilizer. However, when litter-N rates are expressed as an estimate of PAN, the proportion of the total N as PAN for upland crops (50–60%) is usually much greater than for rice (18%).

Assuming that most of the urea N had been taken up by PD, the difference in N uptake between PD and HDG represents mineralized organic N from the soil which may have included some urea-N that was initially immobilized. Between PD and HDG, N uptake by rice receiving no urea, FPL, or PPL ranged from 20 to 32 kg N ha–1, which represents equal or greater N uptake of N from poultry litter by PD. For urea applied at 101 kg N ha–1, up to an additional 15 kg N ha–1 (average among site-years = 4 kg N ha–1) was taken up between PD and HDG. In comparison, total-N uptake for 270 kg total-N ha–1 as FPL and PPL between PD and HDG accounted for up to an additional 55 kg N ha–1 (average among site-years = 15 kg N ha–1). The difference in N uptake between PD and HDG indicates that appreciable amounts of organic N contained in FPL and PPL continued to mineralize after flooding, albeit slower under anaerobic conditions, and may reduce the need for topdressed N at PD. Organic N mineralized after flooding would not undergo nitrification and be susceptible to denitrification.

The efficiency of N uptake by flood-irrigated rice might be improved by altering the time of flooding or the cultural production system. Perhaps establishing the permanent flood earlier than the 5-leaf stage in the direct-seeded system or use of water seeding where the flood is applied immediately after poultry litter is incorporated would improve the recovery of mineralized-organic N. Delaying the time of flooding for PTBS04-DF had no beneficial or detrimental influence on net-N uptake or grain yield compared with rice on the silt-loam soils that were flooded at the 5-leaf stage. Grigg et al. (2000) reported that delaying the establishment of the permanent flood until 20 d after the 5-leaf stage affected rice shoot growth, but did not affect root-length density or grain yields.

Similar to net-N uptake, net yield response to N source also showed that urea was a superior N source compared with FPL and PPL, which did not differ (Table 7 and Fig. 3). However, near maximal yields were sometimes produced with the greatest rates of FPL and PPL indicating that the urea-N equivalents of FPL and PPL were somewhat greater than 18% as indicated by net-N uptake at HDG.

To evaluate this observation, net grain yield data were reanalyzed (as described in the Materials and Methods section) using all N rates for FPL and PPL, but only the rates (≤101 kg urea-N ha–1) which would provide a linear net-yield response for urea-N. As expected, net grain yield increased linearly and significantly (P < 0.0001) and showed common linear slopes within each N source across site-years (Fig. 4A). The slope for urea (43 kg grain kg urea-N–1) was greater than the slopes for FPL (10 kg grain kg total-N–1) and PPL (12 kg grain kg total-N–1), which were statistically similar (data not shown).


Figure 4
View larger version (24K):
[in this window]
[in a new window]
  		Fig. 4. Net rice grain yield as affected by N rate and N source for research trials conducted on five silt loam soils using actual urea-N rates (≤101 kg N ha–1) and fresh and pelleted litter N rates (≤270 kg total-N ha–1) (Fig. 4A) and a comparison of actual urea-N rates (≤101 kg N ha–1) with poultry litter total-N rates adjusted to urea-N fertilizer equivalents (Fig. 4B). For Fig. 4A, mean net grain yields for each site-year are shown as {circ} for urea, {blacktriangleup} for pelleted litter, and {triangledown} for fresh litter.

 
An estimate of the preflood urea-N equivalents of FPL and PPL was determined by the ratio of the slopes for poultry litter (average of FPL and PPL, 11 kg grain kg urea-N–1) and urea-N (43 kg grain kg urea-N–1, Fig. 4A). Each kg total-N ha–1 applied preplant as FPL or PPL produces about 25% of the net grain yield produced by each kg urea-N ha–1 applied preflood (Fig. 4B). If near maximal rice grain yields on silt-loam soils are generally produced by the application of 100 kg urea-N ha–1, 400 kg total-N ha–1 as preplant incorporated FPL or PPL would be needed to produce equal net grain yields on silt-loam soils.

Reasons for the difference between the predicted preflood urea-N equivalence values of poultry litter between net-N uptake and net grain yield are not clearly understood. One possible explanation is that continuous mineralization of organic N from poultry litter may have allowed for additional N uptake after HDG. Rice grain yield increases from topdressed N applications at the late-boot stage or early heading have been documented (Pulley et al., 1999). Net-N uptake by rice has also been shown to increase, decrease, or plateau after heading (Norman et al., 2003).


   SUMMARY
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 SUMMARY
REFERENCES
 
The ultimate goal of this research was to document the preflood urea-N fertilizer equivalents of FPL and PPL so that growers could make the appropriate adjustments to preflood urea-N rates when poultry litter was applied preplant to provide recommended rates of P, K, or both. We also sought to determine whether poultry litter form should be considered in N rate adjustments because the literature contains few field comparisons of N uptake or crop yield between pelleted and fresh forms of poultry litter. Data from these studies with rice suggest that, despite their different physical properties, there is little or no difference in net-N uptake or grain yield between FPL and PPL when applied at equal total-N rates. Although there were no agronomically significant differences between litter forms, there may be advantages and disadvantages of each for use as a nutrient source in row-crop agriculture, especially if the litter must be transported. Poultry litter N-management recommendations for rice do not need to differentiate between fresh and pelleted litter.

Fresh or pelletized poultry litter are not suitable alternative N-fertilizer sources to preflood applied urea for flood-irrigated rice produced in the direct-seeded, delayed-flood cultural system. Five site-years of data on silt-loam soils showed that near maximal rice grain yields were produced only by application of the highest total-N rate (270 kg N ha–1) of FPL and PPL, which also supplies > 70 kg P ha–1. Continuous application of high poultry litter rates would eventually increase soil-test P of soils in eastern Arkansas to high levels or could be detrimental to rice yields (Norman et al., 2003).

Poultry litter applied at 1500 to 2500 kg ha–1 would supply adequate P or K for most silt-loam soils that receive recommendations for P or K and would generally supply 50 to 100 kg total-N ha–1. For this range of application rates, poultry litter would contribute only 7 to 14 kg N ha–1 toward fertilizer-N uptake by HDG. Net-grain yield data indicated that the urea-N equivalents of the total-N in poultry litter was about 25%. Therefore, recommended preflood urea-N rates should be reduced by 25% of the total N applied as poultry litter.


   ACKNOWLEDGMENTS
 
The authors thank the Arkansas Rice Research and Promotion Board for research funding. The authors extend special thanks to Danny Boothe and Shawn Clark for their assistance in establishing and maintaining research plots.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 SUMMARY
 REFERENCES
 
Contribution of the Univ. of Arkansas Agric. Exp. Stn.

Received for publication September 8, 2005.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 SUMMARY
 REFERENCES
 





This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Golden, B. R.
Right arrow Articles by DeLong, R. E.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Golden, B. R.
Right arrow Articles by DeLong, R. E.
Agricola
Right arrow Articles by Golden, B. R.
Right arrow Articles by DeLong, R. E.
Related Collections
Right arrow Nitrogen

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Journal of Natural Resources
and Life Sciences Education		 Vadose Zone Journal
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome