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


     

Published online 8 September 2005
Published in J Environ Qual 34:1896-1909 (2005)
DOI: 10.2134/jeq2004.0413
© 2005 American Society of Agronomy, Crop Science Society of America, and 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 Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (16)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by McGrath, J. M.
Right arrow Articles by Turner, B. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by McGrath, J. M.
Right arrow Articles by Turner, B. L.
Agricola
Right arrow Articles by McGrath, J. M.
Right arrow Articles by Turner, B. L.
Related Collections
Right arrow Animal Waste
Right arrow Nutrients
Right arrow Runoff
Right arrow Nutrient Management
Right arrow Water Pollution

TECHNICAL REPORTS

Waste Management

Broiler Diet Modification and Litter Storage

Impacts on Phosphorus in Litters, Soils, and Runoff

Joshua M. McGratha,*, J. Thomas Simsb, Rory O. Maguirec, William W. Saylorb, C. Roselina Angeld and Benjamin L. Turnere

a Department of Crop and Soil Environmental Sciences, Virginia Polytechnic Institute and State University, Blacksburg, VA 24061
b Department of Animal and Food Science, University of Delaware, Newark, DE 19716
c Department of Soil Science, North Carolina State University, Raleigh, NC 27695
d Department of Animal and Avian Science, University of Maryland, College Park, MD 20742
e Smithsonian Tropical Research Unit, Apartado 0843-03092, Balboa, Ancon, Republic of Panama

* Corresponding author (jmcgrath{at}jhu.edu)

Received for publication October 29, 2004.

   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Modifying broiler diets to mitigate water quality concerns linked to excess phosphorus (P) in regions of intensive broiler production has recently increased. Our goals were to evaluate the effects of dietary modification, using phytase and reduced non-phytate phosphorus (NPP) supplementation, on P speciation in broiler litters, changes in litter P forms during long-term storage, and subsequent impacts of diets on P in runoff from litter-amended soils. Four diets containing two levels of NPP with and without phytase were fed to broilers in a three-flock floor pen study. After removal of the third flock, litters were stored for 440 d at their initial moisture content (MC; 24%) and at a MC of 40%. Litter P fractions and orthophosphate and phytate P concentrations were determined before and after storage. After storage, litters were incorporated with a sandy and silt loam and simulated rainfall was applied. Phytase and reduced dietary NPP significantly reduced litter total P. Reducing dietary NPP decreased water-extractable inorganic phosphorus (IP) and the addition of dietary phytase reduced NaOH- and HCl-extractable organic P in litter, which correlated well with orthophosphate and phytic acid measured by 31P nuclear magnetic resonance (NMR), respectively. Although dry storage caused little change in P speciation, wet storage increased concentrations of water-soluble IP, which increased reactive P in runoff from litter-amended soils. Therefore, diet modification with phytase and reduced NPP could be effective in reducing P additions on a watershed scale. Moreover, efforts to minimize litter MC during storage may reduce the potential for dissolved P losses in runoff.

Abbreviations: DRP, dissolved reactive phosphorus • ICP–OES, inductively coupled plasma–optical emission spectroscopy • IP, inorganic phosphorus • M3-Al, M3-Fe, and M3-P, Mehlich 3–extractable aluminum, iron, and phosphorus, respectively • MC, moisture content • NMR, nuclear magnetic resonance • NPP, non-phytate phosphorus • NRC, National Research Council • OP, organic phosphorus • TDP, total dissolved phosphorus • UMD, University of Maryland • WSP, water-soluble phosphorus


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
PHOSPHORUS LOST FROM agricultural cropland can be a major source of nonpoint pollution of surface and ground waters (Correll, 1998; Maguire and Sims, 2002). Regions of intensive broiler chicken (Gallus gallus domesticus) production often have serious problems with nonpoint P pollution because farm-gate nutrient surpluses, caused by greater import of P in feed to farms than export of P in plant and animal products, can result in overapplication of broiler litter P to soils (Cabrera and Sims, 2000). This can increase soil P concentrations, known to enhance the risk of P loss to water (Beegle et al., 2000; Sims et al., 2000).

One approach to reduce the impacts of land-applied broiler litters on water quality is to modify broiler diets to reduce P excretion. About 50 to 80% of the P in the cereal grains used in broiler diets is phytate, a salt of phytic acid (myo-inositol hexakisphosphate) (Ertl et al., 1998; Ravindran, 1996). However, broilers are monogastric animals that do not possess sufficient amounts of the phytase enzyme needed to hydrolyze nutritionally significant quantities of phytate in the gut of the bird (Morse et al., 1992; Ravindran et al., 1995). Consequently, non-phytate phosphorus (NPP), usually IP in the form of calcium phosphates, is added to poultry feed to supply sufficient digestible P for optimum nutrition and growth. Phytate P, any indigestible mineral P, and P fed in excess of broilers' nutritional requirements are excreted and accumulate in litters. Two approaches proven to reduce P excretion include feeding P closer to the nutritional requirement of the birds and using phytase enzymes to increase the digestibility of phytate P in feed grains (Council for Agricultural Science and Technology, 2002; Ibrahim et al., 1999; Ravindran et al., 1995).

Research has shown that National Research Council (1994) recommendations for NPP concentrations in broiler feed, which provide dietary guidance to broiler producers, may be in excess of actual broiler requirements (Yan et al., 2001; Waldroup et al., 2000). Furthermore, adding phytase to broiler diets, in conjunction with appropriate NPP reductions, has been shown to decrease litter P (Ahmad et al., 2000; Waldroup et al., 2000). Phytase cleaves phosphate from the phytate molecule rendering it available for uptake by monogastric animals. From 30 to 50% of dietary phytate P can be released by phytase supplementation (Angel et al., 2001; Nahm, 2002; Simons et al., 1990; Waldroup et al., 2000), allowing reductions in dietary NPP of 0.1 to 0.2%, and decreasing fecal P concentrations from 35 to 50% (Sims et al., 1999; Yan et al., 2000). Therefore, diet modification can reduce litter P surpluses on farms, decrease overapplication of litter P to cropland, and thereby minimize the buildup of soil P concentrations. As a result, the use of dietary phytase has increasingly been advanced as a water quality best management practice (BMP; Council for Agricultural Science and Technology, 2002; Hatten et al., 2001; Lynch and Caffrey, 1997; Poulson, 2000) and has even been mandated by law in the state of Maryland (Simpson, 1998).

A number of studies have shown that when diets are properly modified with phytase, water-soluble phosphorus (WSP) in poultry manures and litters either decreases or remains unchanged relative to normal diets. Past studies also show that phytase use in diets does not increase the risk of P loss from manured soils. Maguire et al. (2003)(2004) reported that phytase and 25-hydroxycholecalciferol (25-OH D3) additions to poultry diets, where NPP was also reduced, decreased total P and WSP in turkey manures and broiler litters and had no effect on litter WSP to TP ratios, relative to standard diets. When incubated with soil, these litters did not increase soil WSP relative to litters from birds fed NRC diets. Applegate et al. (2003) found total and WSP concentrations were significantly higher in litter from an NRC diet than from diets produced with phytase and reduced NPP. Penn et al. (2004) found turkey diet modification using phytase reduced total P and WSP concentrations in manures and did not increase total P or dissolved reactive phosphorus (DRP) in runoff when surface applied to runoff boxes containing tall fescue. However, Vadas et al. (2004) reported that broiler diets using phytase decreased manure total P, but increased WSP compared to normal diets. On the other hand, they observed no significant impact of phytase use on total or DRP loads in runoff from soil boxes that received surface applications of manure. Miles et al. (2003) reported WSP concentrations in litters from diets using phytase were higher than those using normal diets in litters collected after one flock of broilers, but not in litters from the second and third flocks.

In addition to the above issues, there is increasing concern about the potential for direct P loss from litters stored in or near agricultural fields and the effects of storage practices on P losses from litters after land application. The recently promulgated USEPA concentrated animal feeding operations (CAFO) rule affects large broiler operations (>125000 birds) and as such regulates litter storage (USEPA, 2003). Concerns that direct contact of precipitation with litter stored in fields can lead to contamination of surface and ground waters have caused USEPA to recommend covering stored litters either in a roofed storage facility or with a temporary, impermeable cover such as sheet plastic. Furthermore, questions have been raised over the impact of changes in litter MC during storage on P solubility in the stored litter and the interaction of storage conditions and broiler diet modifications (Waldroup, 2002).

In summary, there is a clear need to reduce P surpluses in areas of intensive animal production and the use of phytase in diets, in conjunction with reductions in NPP, is an effective means to accomplish this. At the same time, there are concerns about the impact of phytase use and storage conditions on P solubility in litters and manures. Therefore, the objective of this study was to evaluate the impacts of using phytase in modified broiler diets and litter storage conditions on total P, P solubility, and the forms of P in litters, litter-amended soils, and runoff.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Broiler Diets and Performance
A floor pen study was conducted using broiler chickens fed six diets: (i) NRC recommendations; (ii) University of Maryland (UMD) recommendations; (iii) UMD with phytase and a 0.064% reduction in NPP; (iv) UMD with phytase, a vitamin D3 metabolite (25-hydroxycholecalciferol [25-OH D3]), and a 0.090% reduction in NPP; (v) NRC with phytase and a 0.1% reduction in NPP; and (vi) UMD with 10% less NPP. During grow-out, broilers were fed four dietary phases: (i) starter = hatch to 18 d; (ii) grower = 18 to 32 d; (iii) finisher = 32 to 42 d; and (iv) withdrawal = 42 to 49 d. Basal diets were analyzed for total P (Heinonen and Lahti, 1981) and phytate P (Rounds and Nielsen, 1993) as modified by Newkirk and Classen (1998), and NPP was determined as the difference between total P and phytate P. Mono-calcium phosphate [Ca(H2PO4)2] was added to achieve target NPP concentrations for each diet. As a result, each diet had the same concentration of phytate P but different concentrations of total P and NPP. Calcium concentrations in diets with reduced levels of NPP were balanced using calcium carbonate and were 0.90, 0.81, 0.71, and 0.61% for starter, grower, finisher, and withdrawal diets. Details on broiler performance will be reported elsewhere; however, on the whole, measured performance criteria (e.g., weight gain, tibia ash, etc.) were similar for all diets (Saylor, unpublished data). Each treatment was replicated nine times except for Treatment 6 which had 10 replicate pens. The pens were 150 by 245 cm and initially contained about 15 cm of wood shavings. Three flocks were grown for 49 d with 56 birds initially placed in each pen per flock. After removal of the third flock, litter samples were collected, analyzed, and then used in a long-term storage study and a rainfall simulation study, as described in the following.

Storage Study Design
The effects of long-term storage on litter properties were assessed using litters generated with four of the six diets fed in the floor pen study (Table 1). Diets selected allowed for direct comparison of the effects of storage conditions, NPP additions, and phytase use on litter properties. Each of the diets had been fed in nine pens for three 49-d flocks. The 49-d survival rate for birds after three flocks was used to select five of the nine pens to obtain litter for use in the storage study. The three pens with the lowest survival rates and the one pen with the highest survival rate were not used.


View this table:
[in this window]
[in a new window]
  		Table 1. Total phosphorus concentrations in the diets used to generate the litters for the storage and rainfall simulation studies and the percentage of phytate phosphorus in those diets.

 
After the birds from the third flock were removed, all litter was collected from each of the five pens selected for the storage study and thoroughly mixed, separately by pen, in a large feed mixer. Two 0.19-m3 trashcans were filled with litter from each of the 20 pens (4 diets x 5 pens, with a total of approximately 45 kg of litter per can) for a total of 40 cans. Two litter moisture treatments were used: (i) litter adjusted to a gravimetric MC (dry basis) of 40%, considered to be the highest MC normally found in Delmarva broiler houses, using deionized water (referred to as wet litter from this point forward) and (ii) litter at its initial MC (approximately 20%, referred to as dry litter), slightly drier than average MC values reported by Sims and Luka-McCafferty (2002) for 200 commercial broiler houses on Delmarva (28–30%). Lids with holes drilled in them were placed on the trashcans to allow for air exchange and prohibit disturbance by pests. The 40 cans were randomly assigned to five replications, with the wet and dry cans from the same pen paired together (four wet and four dry cans in each replication), and stored at ambient temperatures in a broiler house at the University of Delaware Research and Education Center for 440 d. Litter MC was not adjusted during the study but was measured regularly, as described below.

Storage Study Broiler Litter Sample Collection, Preparation, and Analysis
Broiler litter samples (2.5 cm in diameter and the depth of the litter in the can) were collected from cans at 6, 20, 34, 48, 59, 73, 94, 111, 123, 137, 159, 237, 285, 320, 363, 409, and 440 d after moisture adjustment. First and last sample dates were May of 2002 and July of 2003. Litter temperature was recorded at sampling. Temperature was measured in three of the five replications (24 cans). Dial thermometers with 91-cm stems were inserted through holes in the lids to a depth of 76 cm in each can. A high/low thermometer was mounted on a post in the middle of the five replications. Each time litter samples were obtained, ambient air temperature and the high and low temperatures since the last sampling were recorded. Litters stored wet and dry maintained an average MC of 37 and 21%, respectively, during the study (Fig. 1a) . There were no significant interactions between the effects of diet and stored moisture status on litter temperature at any sample date and diet did not significantly affect litter temperature. For the first six months of storage the temperature of litters stored wet was significantly higher than those stored dry (Fig. 1b). Litter MC had little effect on litter temperature during winter months, but wetter litter samples again exhibited a trend for higher temperatures in the summer of 2003. Litter temperatures remained well below values required for thermophilic microorganisms (40°C); hence complete composting should not have occurred (Zibilske, 1997).



View larger version (30K):
[in this window]
[in a new window]
  		Fig. 1. Average and standard deviation of broiler litter (a) moisture content (dry weight basis) and (b) temperature at each sample date for broiler litters stored wet and dry from May 2002 to July 2003. Dashed lines indicate the high and low air temperature since the previous sample date.

 
Litter samples were kept refrigerated and analyzed within 24 h of collection for WSP and MC. Litter WSP was measured by extraction with deionized water (1:10 "as-is" litter to deionized water), shaken for 1 h, centrifuged for 1 h at 1000 x g, and then vacuum-filtered through 0.45-µm filter paper (Millipore, Billerica, MA). Litter pH was measured in deionized water (1:4 litter to deionized water) and litter MC was determined gravimetrically at 60°C. Total P was measured initially and at the end of the study by microwave digestion of 0.5 g of dried, ground (0.5 mm) litter with 7 mL of concentrated HNO3 and 3 mL of H2O2. Total and WSP concentrations in all extracts and digests were determined by inductively coupled plasma–optical emission spectroscopy (ICP–OES). All litter analyses are expressed on a dry weight basis.

Phosphorus chemical fractions were determined on dried and ground litter samples obtained at the outset and conclusion of the storage study. After collection, these samples were dried at 60°C and ground to pass a 2-mm sieve. Subsamples of dried and ground litter from all five replications of each diet–MC treatment were mixed to form a total of eight composite litter samples (four diets x two MC). Since the extraction was performed on composite samples statistical methods could not be applied to determine separation of means. Nonetheless, the results are presented and observed trends are discussed. The P fractionation method of Hedley et al. (1982), as modified by Dou et al. (2000) and Sharpley and Moyer (2000) for manures, was used to sequentially extract P from litters with: (i) deionized H2O; (ii) 0.5 M NaHCO3; (iii) 0.1 M NaOH; and (iv) 1.0 M HCl, using a litter to solution ratio of 1:200 (0.4 g to 80 mL). Litters were shaken for 1 h with deionized H2O and for 16 h with each of the subsequent extractants. After each extraction, each sample was centrifuged at 1000 x g then vacuum-filtered through 0.45-µm Millipore filter paper. All extractions were performed in duplicate. Total dissolved phosphorus (TDP) concentrations in the filtrate were determined by ICP–OES and orthophosphate concentrations (inorganic phosphorus, IP) were determined colorimetrically using a Bran+Luebbe (Delavan, WI) Auto-Analyzer 3 (Murphy and Riley, 1962). Organic phosphorus (OP) in each fraction was calculated as the difference between the TDP concentration and IP concentration.

Solution 31P NMR spectroscopy, using the method of Turner (2004), was conducted to provide structural information on litter P. The same composite litter samples collected before and at the end of the storage study and then used for chemical fractionation of litter P were also used for the 31P NMR analyses. Litters were extracted in triplicate by shaking 1 g of litter with 20 mL of a 0.5 M NaOH and 0.05 M ethylenediaminetetraacetate (EDTA) solution for 4 h. Extracts were centrifuged at 1400 x g for 30 min and 1-mL aliquots were diluted 50-fold and analyzed for total P by ICP–OES. The remaining undiluted solutions from the triplicate extracts were combined and frozen at –70°C, lyophilized, and ground to pass a 500-µm sieve. Immediately before NMR spectroscopy, each freeze-dried extract (approximately 100 mg) was redissolved in 0.9 mL of 1 M NaOH and 0.1 mL of D2O (for signal lock) and transferred to a 5-mm NMR tube. The addition of NaOH ensures consistent chemical shifts and optimum spectral resolution at a solution pH of >13. Solution 31P NMR spectra were obtained using a Bruker (Billerica, MA) Avance DRX 500 MHz spectrometer operating at 202.456 MHz for 31P and 500.134 for 1H. A 5-µs pulse (45°), a delay time of 5.0 s, an acquisition time of 0.8s, and broadband proton decoupling were used for all samples and temperature was maintained at 20°C. Chemical shifts of signals were determined in ppm relative to 85% H3PO4 and assigned to individual P compounds or functional groups based on literature reports (Turner et al., 2003). Signal areas were calculated by integration and P concentrations calculated by multiplying the proportion of the total spectral area assigned to a specific signal by the total P concentration (g P kg–1 dry manure) in the original extract. Litter samples contained primarily orthophosphate and phytic acid; however, quantitatively negligible traces of pyrophosphate and lower-order inositol phosphate esters were detected occasionally.

It is difficult to estimate the error in NMR spectroscopy without acquiring replicate spectra, which is costly and time consuming. However, analytical error for manure and feed samples is estimated at approximately 5% for larger signals and 10% for smaller signals (Leinweber et al., 1997; Kemme et al., 1999).

Rainfall Simulation Study
An indoor rainfall simulation study was conducted using litters collected at the end of the storage study and soils collected from the 0- to 5-cm depths of cropped fields at Chesapeake Farms near Chestertown, MD (Mattapex variant silt loam: fine-loamy, mixed, mesic Aquic Hapludult) and the University of Delaware Research and Education Center in Georgetown, Delaware (Sassafras sandy loam: fine-loamy, siliceous, mesic Typic Hapludult). The rainfall simulation study used methods of the National Phosphorus Research Project indoor runoff protocol (National Phosphorus Research Project, 2001) and evaluated the effects of diet and storage on soil and runoff P when litters were applied at the same total P rate (150 kg P ha–1).

The two soils were sieved (7 mm) and air-dried before amendment with litters collected from the storage study. Fresh litter samples were first collected from each storage study container and analyzed for WSP, total P, and MC to determine litter application rates needed to give the desired total P rate. Subsequently, litters from each replication of each dietary treatment–MC combination were mixed with air-dried, sieved samples of each soil in a small cement mixer and 15 kg of each amended soil were then packed into a runoff box to a depth of 5 cm. The runoff boxes were constructed of plywood treated with varnish and measured 1 by 0.2 m. A control runoff box (no litter) was used for each replication of each soil type. Packed runoff boxes were presaturated and equilibrated for 24 h before runoff. Immediately before applying simulated rainfall, soil cores (3-cm diameter) were taken from the upslope end of the runoff boxes. Holes from soil sampling were filled with amended or control soil and saturated to match surrounding soil. Water-soluble P in all soil samples was determined by shaking 2 g of air-dried, ground, and sieved (2 mm) soil with 20 mL of deionized water for 1 h, followed by centrifugation for 30 min at 1000 x g, and filtration through 0.45-µm Millipore filter paper. Water-soluble P in all filtrates was measured by colorimetric analysis with a Bran+Luebbe Auto-Analyzer 3 (Murphy and Riley, 1962). Soil pH, organic matter by loss on ignition, and Mehlich-3 (M3) P, Fe, and Al (M3: 0.2 M CH3COOH + 0.25 M NH4NO3 + 0.015 M NH4F + 0.013 M HNO3 + 0.001 M EDTA) were determined for the soil samples by standard methods of the University of Delaware (Mehlich, 1984; Sims and Heckendorn, 1991). The M3-P saturation ratio was calculated as the molar ratio of M3-P to [M3-Al + M3-Fe], as described by Sims et al. (2002).

Runoff boxes were placed at a 5% slope and simulated rainfall was applied 24 h after presaturation at 75 mm h–1 for 30 min. The rainfall simulator measured 3.3 m long x 3.3 m wide x 3.4 m high and used a single Tee Jet HH50WSQ nozzle (Spring Systems Co., Wheaton, IL) to produce the desired intensity at greater than 80% uniformity. All runoff was collected, weighed, and sampled for subsequent analysis. For each soil box, the collected runoff was placed in a large beaker and vortexed using a magnetic stir plate. Sediment concentrations were determined by evaporating a 40-mL runoff subsample. Total P concentrations in runoff were determined by microwave-assisted acid digestion of 40 mL of unfiltered runoff and analysis of the digest by ICP–OES (USEPA, 1986). Filtered (0.45-µm Millipore filter paper) runoff was analyzed for dissolved reactive phosphorus (DRP) colorimetrically and for total dissolved phosphorus (TDP) by ICP–OES.

Statistical Analysis
Analyses of variance for all treatment effects in the dietary modification, storage, and rainfall simulation experiments were conducted using the General Linear Model procedure of the Statistical Analysis System, Version 8 (SAS Institute, 1998). Unless otherwise noted the Tukey method was used for all pairwise comparisons. Separation of means was determined to be significant at the 0.05 probability level.


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Phosphorus Concentrations in Broiler Litter
Dietary modification significantly reduced total P in litters generated by three flocks of broilers compared to NRC diets (Table 2). Formulating diets closer to broiler nutritional requirements (UMD) reduced litter total P by 18%, relative to NRC diets, while diets using phytase and in conjunction with decreased NPP further reduced total P by 21% (NRC + phytase) and 29% (UMD + phytase). After storage, total P increased by about 10% in litter stored wet and decreased slightly (approximately 6%) in litter stored dry. Increases in total P during wet storage likely reflect loss of C due to microbial decomposition of litters, expected to be more favorable under wet storage conditions. There was no significant interaction between the main effects of storage moisture and diet, thus dietary trends for total P in litters stored for 440 d were the same as before storage (Table 2).


View this table:
[in this window]
[in a new window]
  		Table 2. Effect of diet and stored moisture status on total phosphorus and water-soluble phosphorus concentrations in broiler litter at the beginning and end of the storage study.

 
Diet had no significant effect on WSP concentrations in litters collected immediately after the third flock of broilers, although there was a trend for lower WSP in litters produced with phytase than those produced without phytase. Additionally, litter WSP for the UMD diets tended to be lower than litters from the NRC diets (Table 2). The dietary trends initially observed in litter WSP remained the same during storage, but were amplified with time and significant differences in WSP due to diet emerged (Table 2). Water-soluble P concentrations increased slightly in litter stored dry and much more dramatically in litter stored wet (Fig. 2) . Greatest increases in WSP in the wet litter occurred between May 2002 (start of the storage) and October 2002, when ambient air and litter temperatures, and thus microbial activity, were highest (approximately 25–35°C; Fig. 1a). During this time WSP concentrations in litter produced by the NRC, NRC + phytase, UMD, and UMD + phytase diets increased 229, 99, 95, and 85%, respectively. Between October 2002 and July 2003 (end of storage) WSP concentrations changed very little, increasing between 5 and 26%. Final WSP concentrations in stored litter from the NRC diets were significantly higher than litters from the UMD diets, regardless of stored moisture status (Table 2). Although not significant, there was a consistent trend toward lower litter WSP concentrations when phytase, in combination with NPP reductions, was used in broiler diets. Our data clearly show that adding phytase to broiler diets, in conjunction with NPP reductions, will not increase litter WSP during long-term storage, even under wet conditions. These results are in agreement with other studies on broiler diets, phytase use, and their impact on excreted P and confirm that excess NPP fed to broilers, especially in the later growth stages, will be excreted and contribute to elevated total and WSP concentrations in litters (Waldroup et al., 2000; Yan et al., 2001).



View larger version (41K):
[in this window]
[in a new window]
  		Fig. 2. Average and standard deviation of water-soluble phosphorus (WSP) concentrations for each diet (NRC, National Research Council; UMD, University of Maryland) in broiler litters stored wet and dry.

 
Chemically Defined Phosphorus Fractions in Broiler Litter
Initial Litter Samples
The effect of modifying broiler diets on chemically defined P fractions in litters could clearly be seen in the initial litter samples (collected immediately after pens were cleaned out; Table 3). In litter from the NRC diet, H2O-extractable P was the largest fraction (31% of litter total P). The NaOH, HCl, and NaHCO3 fractions contained 23, 27, and 7% of total P, while 12% was in the residual fraction (difference between total P and sum of P in chemical fractions). Note that the wider extraction ratio used in the chemical fractionation (1:200) compared to WSP extraction (1:10) resulted in much higher H2O-extractable P (Table 3) than WSP (Table 3).


View this table:
[in this window]
[in a new window]
  		Table 3. Effect of diet{dagger} and storage on total phosphorus in five chemically defined phosphorus fractions and on solution 31P nuclear magnetic resonance (NMR) spectroscopy orthophosphate and phytic acid concentrations in broiler litter.

 
The NRC + phytase diet generated litter that was considerably different than litter produced with the NRC diet (Table 3). All fractions contained lower P concentrations, reflecting the reduction in total P excreted by broilers fed this modified diet. Most notably, the HCl fraction was 62% lower in the NRC + phytase litter than in the NRC litter. While the HCl-P fraction in soils is usually attributed to Ca-P, in manures it is likely to contain significant amounts of recalcitrant organic P, such as phytate P. Maguire et al. (2004), using the same chemical fractionation scheme as our study, showed by solution 31P NMR that all orthophosphate in broiler litters and turkey manures was extracted by H2O and NaHCO3; additional orthophosphate extracted by the HCl step was suggested to originate from acidic hydrolysis of phytate P in litters and manures. Hunger et al. (2005) reported that HCl extracted 10-fold more Ca-P from broiler litters than was measured by solid-state 31P NMR and postulated that HCl mainly extracts phytate P. Therefore, in our study, it seems likely that the decrease in HCl-P in NRC + phytase litters was a result of the hydrolysis of phytate in broiler diets by phytase. This, in turn, shifted P from the HCl fraction to the other three fractions, increasing the percentages of NaOH, NaHCO3, and H2O P extracted (67% of total P) relative to the NRC litter (61% of total P). Additionally, more of the total P was found in the residual pool in the NRC + phytase litter (21%) compared to NRC litter (12%).

Substantially less NPP was fed to the broilers in the UMD diet, compared to the NRC diet, resulting not only in lower total P in the UMD litters but also in changes in P speciation (Table 3). The main causes of the reduction in litter total P with the UMD diet, relative to NRC litters, were decreases in the H2O-P and NaHCO3–P fractions, which together accounted for 38 and 23% of total P in the NRC and UMD litters. The NaOH and HCl pools (3395 and 3117 mg kg–1, respectively) in the UMD litter were comparable in magnitude but more P was in the residual fraction than for NRC litters. This suggests that excess dietary NPP fed as Ca-P in the NRC diet was converted on excretion to highly labile P (H2O-P, NaHCO3–P) in litters. Thus, simply by feeding broilers closer to requirement (e.g., UMD instead of NRC diet), total and labile P in litters can be reduced significantly, in this case by 18% (total P) and 50% ([H2O + NaHCO3]-P).

Addition of phytase to the UMD diet, in conjunction with further reductions in NPP, caused the most significant changes in P speciation and reductions in litter total P compared to the NRC diet. Concentrations of all P fractions were lower in litters from the UMD + phytase diet than in those from the NRC diet. Further, the percentage of total P in the H2O-extractable fraction (23%) of UMD + phytase litters was reduced relative to litter from the NRC (31%) and NRC + phytase diets (34%). We also found that the percentage of HCl-P in litter from the UMD + phytase and NRC + phytase diets was reduced relative to their respective basal diets suggesting enzymatic conversion of dietary phytate P to labile P (Table 3). Consequently, the percentages of [H2O + NaHCO3 + NaOH]-P were 61 and 67% of total P in litter from the UMD + phytase and NRC + phytase diets, compared to 53 and 61% in litters from UMD and NRC diets. Note, however, that the actual concentrations of [H2O + NaHCO3 + NaOH]-P were similar in litter produced by the UMD + phytase (5901 mg kg–1) and UMD diets (5958 mg kg–1) and much lower than values for these three P forms in the NRC (8308 mg kg–1) and NRC + phytase (7091 mg kg–1) diets.

The percentages of total P in inorganic (IP) and organic (OP) forms were also determined in the H2O, NaHCO3, NaOH, and HCl extracts of all litters to provide additional insight into the mechanisms by which modifying diets affected P speciation (Table 4). For the initial samples, from 22 to 41% of total P in these fractions was IP and from 39 to 58% was OP. Litters from NRC diets had higher percentages of IP and those from UMD diets had higher percentages of OP, reflecting the greater percentage of phytate P in UMD diets (Table 1). Other trends observed across all diets were that no OP was detected in the H2O extracts, about half of the NaHCO3–P was organic, and over 90% of the P in the NaOH and HCl fractions was organic. Reducing NPP in diets (UMD and UMD + phytase) decreased IP in the H2O fraction compared to NRC diets (NRC and NRC + phytase) while increasing, by about 10%, the percentage of OP in the NaOH and HCl fractions. Diet had little effect on the percentages of IP or OP in the NaHCO3 fraction. Diets with phytase (NRC + phytase and UMD + phytase diets) had slightly higher percentages of H2O-P (3 and 7%) and lower percentages of HCl-P (11 and 7%) than comparable diets without phytase (NRC and UMD). This provides further evidence that phytase hydrolyzes phytate in diets, converting it to IP that is absorbed by the bird or excreted, resulting in a reduction in OP in the resulting litters.


View this table:
[in this window]
[in a new window]
  		Table 4. Effect of dietary modification and storage conditions on the percentages of inorganic phosphorus (IP) and organic phosphorus (OP) in extracts of each chemical fraction of broiler litters.{dagger}

 
Stored Litter Samples
During storage, considerable changes occurred in the P forms in litter, but these changes were more affected by storage moisture conditions than diet (Tables 3 and 4). When discussing these changes it is important to remember that total P values were about 6% lower for litters stored dry and 10% higher for litters stored wet, relative to initial samples. As noted earlier these differences in total P were most likely due to heterogeneity in the litters and, in the case of the wet litter, loss of C by microbial oxidation during storage. Therefore, while we present the actual P concentrations in each chemical fraction, the effects of diet and storage on P fractionation in litters are better reflected by the changes shown in the percentages of total P in each fraction.

During dry storage, for all diets, the main trends noted for P concentrations were increases in H2O-P and no change or slight decreases in the NaHCO3–P, NaOH-P, and HCl-P. When expressed as a percentage of total P, H2O-P increased from 2% (NRC) to 17% (NRC + phytase), NaHCO3–P decreased by <3%, NaOH-P basically remained unchanged, and HCl-P decreased from 1 to 7% in all diets except NRC + phytase, where it increased by 3% (Table 3). Greater changes occurred in litters stored wet, where H2O-P concentrations increased, relative to initial samples, by 107 (NRC) to 284% (UMD). Thus, in wet litters, H2O-P accounted for 23 to 39% more of total P than in litters analyzed immediately after cleanout. The increase in the H2O-P fraction coincided with decreases in P concentrations and percentages of total P in all of the other three pools. Greatest decreases in extractable P concentrations occurred in the NaOH (50–69% decreases) and HCl (5–47% decreases) fractions. Averaged over all diets, the percentages of total P in the [NaOH + HCl]-P fractions, which likely represent organic P, decreased from 50% in initial samples to 27% in litters stored wet. By comparison, the average percentage of [NaOH + HCl]-P in dry litters was 47%. Examination of the percentage of IP and OP in each fraction of stored litters helps explain changes observed in the forms of litter P. Overall, more IP was extracted from litter stored wet (52–76% of total P) than litter stored dry (29–50%) (Table 4). The major shift in chemical P fractions observed in wet litters, relative to initial and dry litters, regardless of diet, was a marked increase in IP in the H2O fraction, accompanied by decreases in OP in the NaHCO3, NaOH, and HCl fractions (Table 4). These trends are consistent with a greater mineralization of OP to IP by a more active microbial population in wet litters.

Solution Phosphorus-31 Nuclear Magnetic Resonance Characterization: Orthophosphate and Phytic Acid Phosphorus
Initial Litter Samples
Solution 31P NMR spectroscopy of NaOH-EDTA extracts of litter generally supported results of the chemical sequential fractionation of litter P and provided direct evidence for the effects of diet and storage on P speciation (Fig. 3) . Overall, NaOH-EDTA extracted approximately 94% of litter total P (as measured by microwave digestion). The majority of P extracted was determined to be phytic acid and orthophosphate, comprising, on average, 56 and 39% of total P, respectively (Table 3). By comparison, NaOH + HCl (primarily extracts OP) and H2O + NaHCO3 (primarily extracts IP) extracted an average of 50 and 34% of litter total P. Orthophosphate and phytic acid concentrations in the initial, dry, and wet litters, as determined by NMR, were highly correlated with [H2O + NaHCO3]-P and [NaOH + HCl]-P determined by chemical fractionation (Fig. 4) ; these results are in agreement with those of Turner and Leytem (2004). This suggests that the chemical fractionation method we used can accurately identify the major inorganic and organic P species in litters (orthophosphate, phytic acid), of importance because of the costs and limited availability of NMR spectroscopy.



View larger version (17K):
[in this window]
[in a new window]
  		Fig. 3. Solution 31P nuclear magnetic resonance (NMR) spectra of NaOH-EDTA extracts of broiler litter after (a) dry and (b) wet storage from May 2002 to July 2003. NRC, National Research Council; UMD, University of Maryland.

 


View larger version (18K):
[in this window]
[in a new window]
  		Fig. 4. Relationship between [H2O + NaHCO3] inorganic phosphorus (IP) concentrations and orthophosphate P (determined by nuclear magnetic resonance [NMR]) and the relationship between [NaOH + HCl] organic phosphorus (OP) concentrations and phytic acid P (determined by NMR) in stored broiler litter (wet and dry).

 
Litters generated by the NRC diet contained 57% phytic acid and 42% orthophosphate. Adding phytase and reducing NPP (NRC + phytase diet) decreased orthophosphate and phytic acid concentrations by 12 and 32%, compared to the NRC diet. This shows that phytase addition to diets that provide NPP in excess of broiler nutritional requirements (e.g., NRC diet) will decrease labile P forms (e.g., orthophosphate), but not to the same degree that it reduces phytic acid excretion. In the UMD diet, where NPP was fed closer to broiler nutritional requirements, litter orthophosphate concentrations were reduced by 36% compared to the NRC diet (5710 to 3670 mg kg–1) and phytic acid P concentrations decreased by 7% (Table 3). The decrease in litter orthophosphate was likely due to lower concentrations of NPP (e.g., Ca-P) in UMD diets and litters. Supportive evidence for this interpretation is provided by the lower concentrations of P and lower percentages of IP in the H2O and NaHCO3 fractions of UMD litters compared to those from NRC diets (Tables 3 and 4). The reduction in litter phytic acid concentrations suggests that broilers fed the UMD diet were able to hydrolyze phytate P in the corn–soybean meal diet. Past research has shown that this can occur, especially when broilers are fed diets deficient in available P (Van der Klis and Versteegh, 1996). Our findings support the conclusions of others (Yan et al., 2001; Waldroup et al., 2000) that NRC recommendations for NPP exceed broiler requirements. The excess NPP, which is provided as soluble Ca-P, is excreted primarily as labile IP measured in the H2O and NaHCO3 fractions or as orthophosphate measured by 31P NMR. Finally, the UMD + phytase diet had the lowest litter orthophosphate and phytic acid concentrations (44 and 38% of values for NRC litters), again showing the effectiveness of dietary modification at reducing P excretion (Table 3). As with NRC diets, adding phytase to the UMD diet and reducing NPP fed was more effective at reducing litter phytic acid than orthophosphate concentrations (34 and 13% decreases relative to UMD litters).

When expressed as a percentage of total P excreted, the NMR results paralleled the chemical P fractionation data. Reducing dietary NPP (UMD vs. NRC or UMD + phytase vs. NRC + phytase) decreased the percentage of orthophosphate and increased the percentage of phytic acid in litters, similar to changes observed in [H2O-P + NaHCO3–P] and [NaOH-P + HCl-P]. Adding phytase to diets, along with reductions in NPP (NRC vs. NRC + phytase or UMD vs. UMD + phytase), slightly increased or did not change the percentage of orthophosphate and decreased the percentage of litter phytic acid, also similar to changes observed for [H2O-P + NaHCO3–P] and [NaOH-P + HCl-P]. Our results indicate that dietary phytase hydrolyzed phytate in the feed, making it available to the broilers and leading to a decrease in total P excreted.

Finally, of some note is the observation that the percentage of litter phytic acid P from the UMD + phytase diet (50%) was less than the NRC (57%) and UMD (65%) litters and the same as the NRC + phytase litter. This is despite the fact that, averaged over all four growth phases, the UMD + phytase diet had the highest ratio of phytate P to total P in feed (56%) compared to UMD (50%), NRC + phytase (50%), and NRC (43%) diets (Table 1). This again indicates that broilers fed diets that more closely meet their NPP requirements more efficiently utilize phytate P.

Our results are in agreement with others reported in the literature, showing that poultry diets modified with phytase and reduced NPP supplements decreased total P and WSP and did not increase the percentage of total P that was soluble relative to standard diets (Applegate et al., 2003; Maguire et al., 2003, 2004; Penn et al., 2004). In contrast, Vadas et al. (2004) reported that dietary phytase increased WSP concentrations in raw manure collected over the first six weeks of broiler growth. However, the diets and manure collection procedures used in their study differed greatly from ours, which may have contributed to the differences between the results of the two studies.

Stored Litter Samples
As with chemical P fractionation, 31P NMR spectra showed that litter P speciation was more influenced by moisture conditions during storage than by diet (Fig. 4; Table 3). In litters stored dry, trends for the effects of diet on the concentrations and percentages of orthophosphate P and phytic acid were very similar to those observed for initial litters. The main trend continued to be that the percentage of total P as orthophosphate was higher in the NRC and NRC + phytase litters, while the percentage of phytic acid was greater in the UMD and UMD + phytase litters. Feeding closer to requirement (UMD and UMD + phytase vs. NRC and NRC + phytase) decreased orthophosphate and increased phytic acid concentrations and percentages in litters. Conversely, adding phytase to NRC or UMD diets increased the concentrations and percentages of orthophosphate in litters while decreasing phytic acid, further evidence of the efficacy of phytase at hydrolyzing dietary phytate P. As with initial litters, NMR results for litters stored dry were consistent with chemical fractionation trends.

For litters stored wet, much greater increases in orthophosphate concentrations and decreases in phytic acid concentrations were observed compared to initial litters and dry storage (Table 3). The NMR results agreed well with chemical fractionation data. Average values for the percentages of orthophosphate and phytic acid in wet litters, over all diets, were 62 and 34%; in comparison, the average percentages of [H2O + NaHCO3]-P and [NaOH + HCl]-P were 63 and 27%. In general, the chemical fractionation and NMR data clearly indicate that after 440 d of wet storage, the influence of diet on P speciation in litters from all diets decreased and storage conditions became more important. Wet storage resulted in litters mainly dominated by H2O-extractable IP (orthophosphate), as opposed to a predominance of NaOH and HCl OP (phytic acid) in initial litters and those stored under dry conditions (Tables 3 and 4).

Our results are important considering recent efforts to regulate disposal and storage of litter generated in large broiler operations (USEPA, 2003). During storage, litter WSP, averaged across diets and storage MC, increased approximately 65%. Similarly, Vadas et al. (2004) reported that WSP increased about 52% in raw manure that went through freeze–thaw cycles. Chemical fractionation of P and determination by 31P NMR of orthophosphate and phytate concentrations helped to elucidate the mechanisms that increased WSP in the stored litter. During dry storage little change was seen in P forms. However, during wet storage P shifted dramatically from the organic forms extracted by NaHCO3, NaOH, and HCl to H2O-IP. These chemical shifts and the increases in litter temperature that coincided with the greatest fluctuations in WSP showed that microbial activity in stored litter plays an important role in hydrolyzing OP and increasing the labile P pool. Inorganic P comprised between 22 and 41% of the total P extracted by chemical fractionation of the initial litter samples and made up 52 to 66% of the total P extracted after wet storage. The relative increases in IP did not relate to phytase use; therefore, it appears that phytase loses its efficacy after ingestion by the broilers and does not impact P solubility after excretion.

Phosphorus in Litter-Amended Soils and Runoff
The soils used in the rainfall simulation studies were typical of those in the Mid-Atlantic United States. The Sassafras sandy loam was a moderately acidic (pH 6.3), low organic matter (12 g kg–1) soil with a soil test (M3-P) value of 99 mg kg–1, and a P saturation ratio of 0.12. The Mattapex silt loam had a pH of 5.9, organic matter content of 16 g kg–1, and M3-P and P saturation ratio values of 63 mg kg–1 and 0.05. To put the soil P values into perspective, agronomic critical values for M3-P in this region typically range from 30 to 50 mg kg–1, the median M3-P value for agricultural soils in Delaware is about 75 mg kg–1, the Delaware Nutrient Management Commission has mandated P-based nutrient management planning for soils with M3-P values greater than 150 mg P kg–1, and a P saturation ratio of 0.15 has been suggested has an environmental upper limit for Delaware soils (Sims et al., 2002).

In the rainfall simulation studies no statistically significant interactions were observed between diet, litter storage conditions, and soil type. Therefore, the following discussion focuses on the main effects of these factors on soil and runoff P.

Litter applications at equal total P rates (150 kg P ha–1) increased soil WSP, M3-P, and P saturation ratio relative to unamended soils, but no significant differences were found in any form of soil P between litters from different diets (Table 5). A trend did exist for higher soil WSP when NRC and NRC + phytase litters were applied compared to UMD and UMD + phytase litters, expected given the higher amounts of WSP added with NRC litters (Table 5). Litter storage conditions resulted in significant differences in soil WSP, but did not affect M3-P or P saturation ratio. Wet storage of litters increased the percentage of litter H2O-P and orthophosphate compared to dry storage, resulting in greater additions of litter WSP and higher soil WSP (Tables 3 and 5). There was an effect of soil type on soil WSP, M3-P, and P saturation ratio, with all being significantly higher in the Sassafras sandy loam than the Mattapex silt loam (Table 5). This was expected, given that the Sassafras soil was initially higher in all forms of soil P and because of its lower buffering capacity and higher P saturation ratio, was more susceptible to increases in soluble and labile P when litters were applied.


View this table:
[in this window]
[in a new window]
  		Table 5. Main effect of dietary modification and storage on broiler litter application rates (150 kg total P ha–1) and litter water-soluble phosphorus (WSP) added in the rainfall simulation study, and main effect of P application on soil WSP and Mehlich-3 phosphorus (M3-P) concentrations and P saturation ratios and the resulting dissolved reactive, total dissolved, and total P concentrations produced in runoff.

 
Phosphorus concentrations in runoff followed treatment trends for soil P (Table 5). Litter application increased dissolved P (DRP, TDP) concentrations relative to unamended soils. However, although the trend for the effect of diet on runoff P concentrations (NRC > NRC + phytase > UMD + phytase > UMD > control) was consistent with the concentrations and percentages of litter H2O-P and orthophosphate and WSP in litter-amended soils, there were no significant differences in runoff DRP or TDP due to diet. Significantly higher TP concentrations in runoff from soils amended with the NRC diet reflect the slightly higher M3-P and P saturation ratio values for this treatment. Dissolved P concentrations in runoff were significantly greater for soils amended with wet litters, which had higher soil WSP, but soil type did not affect DRP or TDP (Table 5). Higher TP concentrations in runoff from the Mattapex soil reflect higher sediment concentrations in runoff from that fine-textured soil (5.6 g L–1) relative to the Sassafras sandy loam (2.23 g L–1).

In our study, changes in P solubility during storage had a significant impact on P concentrations in soil and runoff. Microbial activity mineralized significantly more organic P in litter stored wet than in litter stored dry, resulting in substantially higher concentrations of P in more labile inorganic forms. In the runoff study, this translated into significantly higher soil WSP concentrations and TDP and DRP concentrations in runoff from soils amended with litter stored at a MC of 40% compared to soils amended with dry litter (MC of 24%) As a result, the benefits attained through dietary modification were overshadowed by changes that occurred during wet storage. The implications of these findings are far-reaching, especially in light of current efforts by the USEPA to regulate litter storage. Managing stored litter to minimize MC increases during storage can significantly reduce dissolved P losses in runoff, even when applied at the same total P rate. As a result, if litter must be stored before land application every effort should be made during production and storage to keep the litter as dry as possible to limit losses of dissolved P, which is immediately available for algal uptake after entering surface waters.

The results of this study agree with others that incorporation of P sources results in relatively rapid sorption of litter WSP that tends to minimize differences in soil WSP due to amendment type (Maguire et al., 2004; Kleinman and Sharpley, 2003; Kleinman et al., 2002). Therefore, when broiler litter is applied and incorporated at the same total P rate, slight increases in WSP concentrations in broiler litter caused by diet modification will have little impact on soil and runoff P concentrations over the long-term. However, in the current study there was a strong relationship between soil WSP and DRP (Chesapeake Farms: r2 = 0.93, significant at the 0.001 probability level; Georgetown: r2 = 0.76, significant at the 0.01 probability level; Fig. 5) indicating that continued application of P to these soils beyond crop requirements will result in increased P losses in runoff. Therefore even with broiler diet modification proper management of P application is required to minimize P losses from agricultural fields. Furthermore, our results are in agreement with those of Penn et al. (2004) that phytase use in poultry diets does not increase total P or DRP concentrations in runoff.



View larger version (18K):
[in this window]
[in a new window]
  		Fig. 5. Effect of broiler litter (BL) application at 150 kg P ha–1 on the relationship between soil water-soluble phosphorus (WSP) concentrations and runoff dissolved reactive phosphorus (DRP) concentrations generated by simulated rainfall.

 

   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
We evaluated the effect of modifying broiler diets, using phytase and reduced NPP concentrations, and litter storage on P concentrations and forms in the broiler litters, litter-amended soils, and runoff. Reducing NPP supplementation to closer meet broiler nutritional requirements along with use of phytase was effective at reducing litter total and WSP. Storing litter at a MC of 40% compared to storage at its initial MC (approximately 24%) significantly increased litter WSP regardless of diet. Evaluation of litter P speciation by sequential chemical fractionation and 31P NMR showed that excess NPP in broiler diets is primarily excreted as orthophosphate or inorganic, H2O-extractable P. Further, these techniques confirmed that dietary phytase hydrolyzes phytate in the broiler making it available for absorption during digestion, and leading to decreases in the amount of phytate excreted without appreciably increasing the amount of labile P excreted. Increases in the labile P pool seen during wet storage were not related to diet; therefore, it appears unlikely that phytase continues to hydrolyze phytate during litter storage. Rainfall simulation study results showed that, when applied at the same total P rate, diet did not influence P losses in runoff. With respect to our study, the UMD diet with added phytase would produce the most benefit at the watershed scale by reducing P import and decreasing the amount of WSP generated in litter. Furthermore, proper management of the litter during production and storage to minimize increases in MC could significantly limit increases in soluble P forms, thereby reducing the potential for P losses from litter-amended soils.


   ACKNOWLEDGMENTS
 
This project was funded, in part, by the USDA grant program "Initiative for Future Agriculture and Food Systems" (IFAFS Grant 00-52103-9700).


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES