|
|
||||||||
a Eastern Shore Agricultural Research and Extension Center, Virginia Polytechnic Institute and State Univ., 33446 Research Dr., Painter, VA 23420
b Dep. of Agronomy and Horticulture, New Mexico State Univ., Room 127N, Skeen Hall, P.O. Box 30003, Las Cruces, NM 88003-8003
c Dep. of Crop and Soil Environmental Sciences, Virginia Polytechnic and State Univ., 330 Smyth Hall, Blacksburg, VA 24061
* Corresponding author (jwarren{at}ars.usda.gov)
| ABSTRACT |
|---|
|
|
|---|
Abbreviations: ATPL, Poultry litter treated with alum CR, Phosphorus application rate of 35 kg P ha1, equivalent to estimated crop P removal 2CR, Phosphorus application rate of 70 kg P ha1, equivalent to estimated 2 yr of crop P removal, applied before planting in 2000 and 2002 NPL, Non-treated poultry litter without alum additions PAN, Plant available N Pi, Inorganic P Po, Organic P TSP, Triple superphosphate fertilizer
| INTRODUCTION |
|---|
|
|
|---|
Historic applications of animal manures, such as poultry litter, at rates in excess of crop removal have resulted in elevated concentrations of P in agricultural soils (Kingery et al., 1994; Sharpley et al., 1993), specifically in those soils proximate to concentrated animal production facilities that have received long-term applications of P in animal wastes. These increases in soil P can correspond to increased concentrations of P in surface water runoff (Sharpley, 1995; Pote et al., 1996). Also, the timing, method and rate of P application can have profound impacts on the amount of P lost from agricultural land receiving poultry litter applications (Sharpley et al., 1994). Therefore, land managers are currently encouraged through regulatory or cost share programs to decrease litter application rates and incorporate conservation tillage into their land management strategies. However, decreased litter application rates increase the land base required for utilization/disposal of poultry litter. In areas of intensive poultry production, transportation costs associated with dispersing litter to lands located farther from production facilities can become prohibitive. In addition, conservation tillage practices such as no-till reduce P losses associated with litter particle loss and soil erosion; however, they have the potential of increasing soluble P losses (Sharpley et al., 1994). An alternative management strategy is chemical alteration of poultry litter to reduce soluble P concentrations.
Various chemical amendments have been evaluated as to their effectiveness in reducing soluble P concentrations in poultry litter. Many of these amendments have included Ca, Al, and Fe containing compounds such as alum [Al2(SO4)3·16H2O], ferrous sulfate (FeSO4·7H2O) (Shreve et al., 1996), sodium aluminate (Na2Al2O4), quicklime (CaO), slaked lime [Ca(OH)2], calcitic limestone (CaCO3), dolomitic limestone [CaMg(CO3)], gypsum (CaSO4·2H2O), ferrous chloride (FeCl2·4H2O), ferric chloride (FeCl3), and ferric sulfate [Fe2(SO4)3·2H2O] (Moore and Miller, 1994). Of these amendments, alum has been found to be one of the most cost-effective options (Moore et al., 1999). Initial laboratory evaluation of alum as a poultry litter amendment revealed that additions of alum at rates of 100 g kg1 could reduce the concentration of soluble P by 75% (Moore and Miller, 1994; Shreve et al., 1996). A subsequent large farm-scale evaluation of alum as a litter amendment confirmed these findings (Sims and Luka-McCafferty, 2002). Sims and Luka-McCafferty (2002) applied alum to 97 houses at a rate of 0.09 kg alum m2 flock1. This rate was estimated to be equivalent to that required to result in a final litter containing 10% alum by weight if previous litter was not present in the house at the time of application. However, the houses included in this study varied widely in the amount of litter present at the initiation of the study. Despite this variation in litter thickness, applications of alum resulted in an average decrease in soluble P of 72% compared with that in litter collected from 97 poultry houses not receiving the alum treatment.
Various studies have focused on evaluating the mechanisms by which this reduction in P solubility resulting from alum treatment of poultry litter is achieved. X-ray adsorption near edge structure spectroscopy (XANES) analysis was used to determine that in NPL, P was present as weakly bound inorganic and organic P with some Ca-phosphate compounds present (Peak et al., 2002). Analysis of ATPL suggested that the alum applied to the litter precipitates as amorphous Al(OH)3 to which inorganic P was adsorbed. Further investigation into the speciation of P in poultry litter using solid-state 31P nuclear magnetic resonance (NMR) spectroscopic analysis found the presence of a complex mixture of organic and inorganic orthophosphates (Hunger et al., 2004). This analysis again revealed a Ca-P phase, which the researchers attributed to a surface precipitate on CaCO3 in both ATPL and NPL. They found P associated with Al only in the ATPL and stated that this was likely a mixture of poorly ordered wavellite and surface complexes with Al(OH)3.
This reduction in the soluble P concentration in ATPL has been shown to result in decreased runoff P concentrations of soluble P following surface application. Smith et al. (2004) found that ATPL applications resulted in 47 to 74% reductions in soluble P concentrations in runoff compared with NPL depending on dietary treatment. Warren et al. (2006) found that N-based applications of ATPL for conventionally tilled corn (Zea mays L.) production resulted in 61% reductions in runoff concentrations of soluble P compared with equivalent applications of NPL. Alum-treated poultry litter applications also resulted in 63 to 87% reductions in P concentrations in runoff compared with NPL when applied to fescue pasture (Shreve et al., 1995).
Despite this collection of data evaluating the use of alum as a poultry litter amendment, there is only limited data evaluating the effects of long-term applications of ATPL on the solubility and forms of P in soils. Initial laboratory incubation evaluations found that after 294 d the soluble P concentrations in soils treated with ATPL were significantly reduced compared with those found in soils treated with NPL (Shreve et al., 1996). Warren et al. (2006) found that N-based applications of ATPL to field corn resulted in lower soil P status, as determined by Mehlich-1 extraction, compared with that found in treatments receiving equivalent N-based applications of NPL both in the Coastal Plain and Piedmont regions of Virginia. In Coastal Plain soils, the reduced P status could be attributed to reduced P application rates, and also to decreased availability of the applied P. Moore et al. (1998) presented soil data collected after 3 yr of ATPL application to tall fescue, which showed lower concentrations of both Mehlich-3 and H2O extractable P in soils receiving ATPL compared with those receiving NPL even when both were applied at similar P rates.
Data evaluating speciation of P in soils after applications of ATPL were presented by Staats et al. (2004) who incubated three soils with ATPL and NPL for 25 d and then performed the sequential fractionation of Chang and Jackson (1957) on the soils. They found little difference in the distribution of P among soils treated with ATPL compared with those treated with NPL. In fact, the only significant difference among the litter sources was found in the NH4F extraction, in which P concentrations were elevated in soils receiving ATPL compared with those receiving NPL. Elevated P in NH4F extractions likely resulted from the increased association of P with Al in soils treated with the ATPL. Despite this limited laboratory incubation data evaluating P speciation in soils treated with ATPL, nothing is known about the long-term effects of ATPL applications on the distribution of P in soils under crop production.
The objectives of this study were to evaluate the impacts of ATPL applications on tall fescue hay production and the distribution of P in soils. The responses of forage yield and P removal to ATPL applications based on current litter management strategies were evaluated. Also, changes in the chemical distribution of soil P resulting from these applications of ATPL were evaluated using sequential fractionation (Tiessen and Moir, 1993).
| MATERIALS AND METHODS |
|---|
|
|
|---|
|
Phosphorus sources used in this study consisted of commercial phosphate fertilizer {TSP [Ca(H2PO4)2·H2O]}, ATPL, and NPL, applied at four different P rates (Table 2). In addition to a no-P check treatment, P rates were based on: (1) applying NPL at rates to meet the N needs of the crop (N-based, NPL); (2) applying ATPL at rates to meet the N needs of the crop (N-based, ATPL); (3) annual estimated crop removal of P (CR); (4) estimated 2-yr crop removal of P (2CR) and (5) soil test P recommendations. All treatments received 112 kg of plant available N (PAN) ha1 as poultry litter and/or NH4NO3. However, the 2CR P application rate did result in PAN application rates in excess of this target N rate for the application of ATPL in 2000 and for both litter sources in 2002 (Table 2). Also the 2CR treatments received P only in 2000 and 2002; in 2001 and 2003 these treatments received only 112 kg N ha1 as NH4NO3. The 112 kg PAN ha1 application rate was chosen to provide sufficient N for an expected yield of 9.0 Mg ha1 assuming a requirement of 12.5 kg PAN Mg1 of forage (Virginia Department of Conservation and Recreation, 2005). The availability of N in the NPL and ATPL was estimated using guidelines developed by the Virginia Department of Conservation and Recreation (2005). These guidelines assume that 50% of the NH4N and 60% of the organic-N is available as PAN for crop uptake in the year of application when litter is surface applied without incorporation. The application rates of litter, PAN, NH4NO3N, and P resulting from the treatment structure are outlined in Table 2. These treatments and supplemental N were broadcast-applied in late March of each growing season because spring growth is most rapid although fescue can continue to grow throughout the year in this area. Supplemental K fertilizer was also applied in late March as fertilizer grade KCl to treatments at rates determined by Virginia Cooperative Extension soil test recommendations to ensure adequate K availability for all treatments.
|
Poultry litter used in this fescue experiment was collected from two different farm scale studies evaluating alum additions to the floor of poultry production houses. Poultry litter applied in the spring of 2000 was collected from two poultry houses included in the study presented by Sims and Luka-McCafferty (2002). In this study one of the house received alum at an approximate rate of 0.09 kg bird1 before the introduction of each flock. The second house received no alum additions. Poultry litter applied in the spring of 2001 through 2003 was collected from two poultry houses located side by side under similar management in the Shenandoah Valley of Virginia. Here the house from which ATPL was collected received alum at an approximate rate of 0.05 kg bird1 before the introduction of each flock. The second house received no alum additions. Litter was collected at the end of a grow-out and stored under cover until application. The average elemental compositions of the litters are presented in Table 3. All elemental analyses were determined on samples by commercial laboratories after drying at 80° and grinding except for NH4N analysis. Ammonium N was determined by distillation and titration after extraction of the litter with a 1:10 litter to 2 M KCl extraction. Total N was determined by Kjeldahl digestion. The remaining elemental analysis was achieved by inductively coupled plasma argon emission spectroscopy (ICPAES) after nitric acid and peroxide digestion (Hoskins, 2003).
|
Composite soil samples were collected annually during the first 2 wk of March to a depth of 10 cm before treatment applications. Soil samples were air dried and ground to pass a 2-mm sieve. Soils were then analyzed for Mehlich-1 soil test P, exchangeable Al (1:5 soil/1 M KCl ratio) (Bertsch and Bloom, 1996), and pH (1:1 soil/water ratio) (Thomas, 1996). In addition, samples collected on 3 Mar. 2004 were analyzed for ammonium oxalate extractable P, Fe, and Al (Pote et al., 1999) and subjected to a P fractionation procedure modified from Tiessen and Moir (1993). The ammonium oxalate extraction was accomplished by combining 0.5 g of soil and 20 mL of 0.2 M ammonium oxalate in a 50-mL centrifuge tube and shaking the solution for 2 h in the dark. This solution was then filtered in the dark through a Whatman #42 filter and analyzed for P, Fe, and Al using ICPAES. The first extraction of the fractionation procedure was accomplished by weighing 0.5 g of soil into a 50-mL centrifuge tube, adding 30 mL of deionized H2O, and shaking overnight (16 h). The suspension was centrifuged at 1200 x g (2000 rpm) for 30 min. The supernatant was then filtered through a 0.45-µm filter and the particles on the filter were washed back into the centrifuge tube using 0.5 M NaHCO3. Additional 0.5 M NaHCO3 was added to bring the final volume to 30-mL and this suspension was shaken overnight. This protocol of extraction was used for the 0.5 M NaHCO3 and also performed for the 0.1 M NaOH and 1.0 M HCl. The remaining soil residue was mixed with 10 mL of concentrated HCl and heated at 80°C in a water bath for 10 min. After removal from the water bath an additional 5-mL of concentrated HCl was added and the mixture was allowed to cool. This suspension was centrifuged at 1200 x g for 30 min and the supernatant decanted without filtration. The remaining soil residue was transferred to 75-mL digestion tubes using deionized H2O. The soil residue was digested using a mixture of H2SO4 and H2O2 at 360°C. Total P was determined on each of the above described extractions using ICPAES. Inorganic P (Pi) was determined on the H2O, NaHCO3, and NaOH extractions after acidification and on the concentrated HCl extraction after neutralization using the Molybdate blue method of Murphy and Riley (1962). The organic P (Po) was calculated as the difference between the total P and Pi.
Analysis of variance and contrast analysis were performed using the SAS PROC GLM procedure (SAS Institute, 2001), to determine significant treatment effects on measured response variables. Fisher's protected LSD was used to separate treatment means. Regression analyses were conducted using the SAS PROC REG procedure.
| RESULTS AND DISCUSSION |
|---|
|
|
|---|
|
|
Soil Phosphorus Status
Mehlich-1 extractable P concentrations in soils collected in the spring of 2004 were elevated in all treatments compared with that found in the 0-P check treatments (Table 5). Also, the N-based treatments resulted in significantly increased Mehlich-1 extractable P compared with the 2CR and CR P application rates. No significant differences in Mehlich-1 extractable P were found among treatments receiving ATPL and NPL.
These results are inconsistent with previous research that reported consistently lower soil test P in soils receiving ATPL (Warren et al., 2006; Moore et al., 1999). Warren et al. (2006) found that each kg of net P ha1 applied as NPL to a Davidson loam soil increased Mehlich-1 P by 0.041 mg P kg1. In contrast, each kg of net P ha1 applied as ATPL increased Mehlich-1 P by 0.031 mg P kg1. In the fescue study, the N-based applications of NPL and ATPL resulted in increases in Mehlich-1 extractable P at rates of 0.055 and 0.057 mg P kg1 of soil per kg net P ha1 applied, respectively. These increases per unit of net P applied are slightly higher than that found at the CR application rate that averaged across source was 0.052 mg P kg1 of soil per kg net P ha1. The inconsistencies between data presented here and in previous research may be explained by the smaller difference between the net P applied as ATPL and NPL in this study (Table 5) compared with that applied to corn by Warren et al. (2006). Net applications of P at N-based rates during the 3-yr corn trial conducted by Warren et al. (2006) totaled 283 and 164 kg P ha1 for NPL and ATPL, respectively, whereas during this 4-yr fescue study net P applications were 181 and 145 kg P ha1 for NPL and ATPL, respectively (Table 5). The lower P application rates specifically in the NPL treatments, compared with those presented by Warren et al. (2006) do not appear to have been sufficient to result in significant differences in the Mehlich-1 extractable P found between treatments receiving the two litter sources at N-based rates. Also the Al to P ratio for ATPL used in this study was lower at 0.5 compared with that used by Moore et al. (1999), which had an Al to P ratio of approximately 1.0.
Ammonium oxalate-extractable P was elevated in the N-based treatments compared with the treatments receiving P at the CR and 2CR rates (Table 5). In addition ammonium oxalate-extractable P was significantly increased in soils receiving ATPL compared with that found in soils receiving NPL (Table 5). However, no significant differences were found when comparing TSP to either of the litter P sources. These are unique data since currently no published data exist that evaluate the effects of ATPL applications on ammonium oxalate-extractable P in soils. A reasonable explanation for the elevated ammonium oxalate-extractable P in soils receiving ATPL is that the treatment of litter with alum results in the formation of Al-P complexes. These complexes are likely amorphous Al-oxides with adsorbed or occluded P (Hunger et al., 2004; Peak et al., 2002). Because P removal during the 4-yr study was similar for ATPL and NPL treatments (Table 5), it can be hypothesized that plant P uptake preferentially utilized P forms that were more available than those associated with the amorphous Al complexes formed in ATPL. Therefore, these Al-P forms apparently accumulated in the surface of soil treated with ATPL and are extractable with ammonium oxalate. It should be noted that the applications of ATPL, which resulted in a 4-yr cumulative Al application of 106 kg Al ha1, did not significantly change the ammonium oxalate-extractable Al (Table 5). This is expected since the average ammonium oxalate-extractable Al in the 10-cm depth of soil sampled would be 2507 kg ha1 assuming that the bulk density of the Davidson soil is 1.4 g cm3. Therefore the Al applied in the ATPL is <5% of the ammonium oxalate-extractable Al inherently present in the soil.
Phosphorus Fractionation
The concentrations of Pi extracted using deionized H2O, 0.5 M NaHCO3, and 0.1 M NaOH from soils collected in the spring of 2004 were significantly affected by treatment (Table 6). No significant differences were found in the 1.0 M HCl extractable P, which had an average concentration among treatments of 44 mg P kg1. Also, there were no significant treatment effects on the concentrations of Pi and Po found in the concentrated HCl extractions nor the concentrations of P found in the H2SO4 digestions. The average concentrations of Pi and Po in the concentrated HCl extraction were 181 and 41 mg P kg1, respectively. The average concentration of P found in the H2SO4 digestion was 233 mg P kg1. These results are similar to those found by Blake et al. (2003) who used a similar fractionation procedure to evaluate soil P forms after long-term P applications and removal. They found that 68 to 89% of the changes in soil P after P applications were in the resin (highly labile), 0.5 M NaHCO3, and 0.1 M NaOH extractions combined with limited changes in the acid-extractable fractions. However, sequential fractionation of NPL has shown that 25% of the total P found in the litter was in the 1.0 M HCl extraction and <2% was present as residual P extracted by hot acid digestion (Dou et al., 2000). Furthermore, Hunger et al. (2005) used the same fractionation procedure on ATPL and found that 37 to 39% of P in the litter was extracted using 1.0 M HCl extraction. However, the soil fractions presented here indicate that litter applications to the Davidson soil were not sufficient to increase the acid-soluble soil P, as is evident by the lack of a significant increase in the 1.0 M HCl extractable fraction compared with the 0-P check treatment.
|
The concentrations of Pi found in the 0.5 M NaHCO3 extracts were elevated in all treatments receiving P applications compared with the 0-P check treatment (Table 6). No significant differences among the CR and 2CR P treatments were found. However, each of the N-based treatments elevated this fraction compared with the CR and 2CR P treatments (Table 6). Within the N-based application rates, both TSP treatments significantly increased 0.5 M NaHCO3extractable Pi compared with that found in the two litter treatments. Also, the N-based NPL application resulted in a significant increase in the 0.5 M NaHCO3 extractable Pi compared with that found in the equivalent ATPL treatment.
The concentrations of Pi extracted using 0.1 M NaOH were not significantly increased by the CR or 2CR P application of any source compared with the 0-P check treatment (Table 6). However, each of the N-based treatments significantly increased the 0.1 M NaOH extractable Pi compared with the 0-P check treatment. Within the N-based treatments, the application of TSP at the rate equivalent to that supplied by the NPL significantly increased the 0.1 M NaOH extractable Pi compared with the remaining N-based treatments with no other significant differences among sources.
In addition to these changes in Pi, significant differences were found in the Po extracted with 0.1 M NaOH (Table 6). Applications of ATPL resulted in significant increases in the concentrations of Po extracted with 0.1 M NaOH compared with the 0-P check treatment, regardless of rate. However, applications of TSP and NPL did not increase this fraction of P above that found in the 0-P check treatment and this fraction is consistently lower in these treatments compared with ATPL treatments.
Although limited data are available on the effects of ATPL on soil P fractions, litter fractionation studies are available for comparison. Dou et al. (2003) evaluated the effects of adding 100 g alum kg1 of litter on the sequential fractionation of P in litter. They found that alum additions resulted in decreases in the H2O extractable and 1.0 M HCl extractable Pi with corresponding increases in Pi extracted using NaHCO3 and 0.1 M NaOH. In addition, the authors noted much higher concentrations of Po in the 0.1 M NaOH extraction of ATPL compared with that found in NPL. The authors attributed this increase in Po to the sorption-entrapment of organic P on Al(OH)3 flocs, which were said to form as a result of alum addition to poultry litter. This helps to explain the elevated 0.1 M NaOH extractable Po found in soils treated with ATPL. The elevated concentrations of this fraction in both ATPL, as demonstrated by Dou et al. (2003) and in soils receiving ATPL, suggest that this 0.1 M NaOH extractable Po may contain P compounds that are stable in the soil environment. The fact that this fraction is elevated not only in the N-based ATPL treatment but also in the 2CR ATPL treatment, which had not received litter applications for 2 yr before the spring 2004 soil collection, suggests that the 0.1 M NaOH extractable Po fraction is relatively stable in soils. Of course there is likely a mechanism by which the accumulation of this fraction in soils receiving ATPL is regulated. The lack of a significant increase in the 0.1 M NaOH-extractable Po found in the treatment receiving N-based applications of ATPL compared with that found in treatments receiving ATPL at the CR and 2CR application rates supports the presence of this regulatory mechanism. This mechanism may be related to the cycling of organic matter within the soil system. Future research may focus on whether such a mechanism exists as well as focus on the persistence of this soil P fraction.
The lack of significant difference in the concentrations of 0.1 M NaOH extractable Pi when comparing ATPL and NPL treated soils can be attributed to the mineral character of the soil. As mentioned previously, the amount of Al added as ATPL will at best increase the amorphous Al content of the soil by <5%. Therefore, the inherent mineral character of this soil will play a greater role in controlling 0.1 M NaOH-extractable Pi, which is described as stable Pi associated with Fe and Al (Hedley et al., 1982), than will the addition of ATPL.
It should also be noted again that ammonium oxalate-extractable P was elevated in soils receiving ATPL treatments. In fact when regression analysis was used to correlate ammonium oxalate P with individual P fractions or combinations of the P fractions, correlation coefficients were optimized when comparing ammonium oxalate P to the summed concentration of Pi found in the H2O, 0.5 M NaHCO3, and 0.1 M NaOH extractions plus the Po extracted using 0.1 M NaOH (Fig. 1 ). This demonstrates that ammonium oxalate extractable P is associated with not only the biologically available Pi extracted using H2O and 0.5 M NaHCO3 and the more stable Pi associated with amorphous material extracted with 0.1 M NaOH (Hedley et al., 1982), but also Po extracted with 0.1 M NaOH. The comparisons of ammonium oxalate-extractable P to the fractionation data suggest that the elevated concentrations of 0.1 M NaOH-extractable Po may account for elevated ammonium oxalate-extractable P concentrations found in soils treated with ATPL.
|
|
| CONCLUSIONS |
|---|
|
|
|---|
The decreases in H2O-extractable and 0.5 M NaHCO3extractable Pi as a result of ATPL applications were accompanied by increases in the concentrations of Po extracted using 0.1 M NaOH. Unlike the suppressed concentrations of Pi extracted using H2O and 0.5 M NaHCO3, which only occurred in the N-based application of ATPL, the increase in 0.1 M NaOH-extractable Po occurred at each application rate. The data show that this P fraction is persistent after ATPL application to soil as evident by the elevated concentration of this P fraction in the 2CR treatments that had not received ATPL for 2 yr before sampling soils in spring 2004. Although the 0.1 M NaOH-extractable Po does appear to persist in soil treated with ATPL, it did not increase as the rate of ATPL application increased, which suggests the presence of mechanisms that limit the accumulation of 0.1 M NaOH-extractable P in soils.
Future research may focus on identifying the P forms found in this fraction and evaluating the mechanisms that control the accumulation of this P fraction in soils receiving ATPL. This will allow for better understanding of the mechanisms by which alum treatment of poultry litter affects soil P distribution. In addition research may focus on the role of SO42 competition for P adsorption sites when applied in ATPL. Also, the impact of liming soils that receive ATPL is of interest. The data presented here demonstrated that ATPL is effective in reducing H2O extractable P in soils limed to a pH near neutral. However, this study cannot be used to assess how lime applications specifically impact P availability. This information is needed to ensure that the treatment of poultry litter with alum is an effective management practice for long-term reductions in the transport of P from agricultural systems.
| ACKNOWLEDGMENTS |
|---|
Received for publication January 5, 2006.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J. G. Warren, C. J. Penn, J. M. McGrath, and K. Sistani The Impact of Alum Addition on Organic P Transformations in Poultry Litter and Litter-Amended Soil J. Environ. Qual., March 1, 2008; 37(2): 469 - 476. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 | |||