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


     

Published online 12 March 2007
Published in Soil Sci Soc Am J 71:551-560 (2007)
DOI: 10.2136/sssaj2006.0253
© 2007 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 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 HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Shober, A. L.
Right arrow Articles by Sims, J. T.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Shober, A. L.
Right arrow Articles by Sims, J. T.
Agricola
Right arrow Articles by Shober, A. L.
Right arrow Articles by Sims, J. T.
Related Collections
Right arrow Phosphorus
Right arrow Nutrient Management

NUTRIENT MANAGEMENT & SOIL & PLANT ANALYSIS

Integrating Phosphorus Source and Soil Properties into Risk Assessments for Phosphorus Loss

Amy L. Shober*,a and J. Thomas Simsb

a Dep. of Soil and Water Science, Univ. of Florida, Gulf Coast Research and Education Center, Wimauma, FL 33598
b Dep. of Plant and Soil Sciences Univ. of Delaware Newark, DE 19716

* Corresponding author (alshober{at}ufl.edu).


   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 IMPLICATIONS AND CONCLUSIONS
 REFERENCES
 
Differences in P solubility of biosolids and manures, as a result of chemical treatment or other factors, have prompted states to incorporate weighting factors, sometimes called phosphorus source coefficients (PSCs), into P risk assessment tools. The value of PSCs in risk assessment has been demonstrated when organic P sources are surface-applied, but not when P sources are incorporated into soils. Ten organic P sources were incorporated into eight Mid-Atlantic soils (0–5 cm) at 135 kg total P ha–1 and incubated for 2, 30, and 180 d. Samples were analyzed for Mehlich 3 phosphorus (M3-P), M3-P saturation ratio (M3-PSR), and water-soluble phosphorus (WSP). Average increases in M3-P (26 mg kg–1) and M3-PSR (0.03) for all soils were similar for untreated manures and inorganic P compared to alum-treated poultry litter (APL) and biosolids (24 mg kg–1, 0.03). In contrast, average concentrations of soil WSP (2 d) were highest following incorporation of untreated manures and inorganic P (5.2–8.0 mg kg–1), followed by biosolids or APL (2.9–3.4 mg kg–1) and unamended soils (2.7 mg kg–1) with similar trends at 30 and 180 d. Source effects were most pronounced for soils with higher M3-PSR. Regression equations using soil M3-PSR and default PSCs could predict soil WSP at 2, 30, and 180 d (r2 = 0.67, 0.69, 0.60, all significant at P = 0.001). Based on our results, we developed a new weighting factor (using M3-PSR and default PSCs) for risk assessment tools that will better predict the potential for P loss when biosolids or manures are incorporated into Mid-Atlantic soils.

Abbreviations: BPR, biological phosphorus removal • DPS, degree of phosphorus saturation • DRP, dissolved reactive phosphorus • ICP-AES, inductively coupled plasma atomic emission spectroscopy • M3-P, Mehlich-3 phosphorus • M3-PSR, Mehlich-3 phosphorus saturation ratio • PSI, phosphorus site index • PSC, phosphorus source coefficient • TP, total phosphorus • WSP, water-soluble phosphorus


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 IMPLICATIONS AND CONCLUSIONS
 REFERENCES
 
Previous research has linked the degradation of water quality to increased losses of dissolved and particulate P from excessively fertilized agricultural soils (Parry, 1998; Sims et al., 1998; Tunney et al., 1997). As a result, national and state guidelines and regulations have been adopted in recent years in the USA that restrict land application of manures and biosolids based on some measure of the risk of P loss to water (Simpson, 1998; Sims, 1999; USEPA, 2003). Most states use the P site index (PSI), a risk assessment tool combining site properties that affect P transport with soil P status and P source management practices to determine when P-based management is needed (Lemunyon and Gilbert, 1993; Sharpley et al., 2003). Currently, most PSIs only consider the total amount of P added during land application of manures or biosolids when calculating the potential for P loss to water (Coale et al., 2002). Previous research has demonstrated, however, that fertilizers, manures, and biosolids vary widely in P solubility and thus have different relative risks of P loss when applied to soils (Kleinman et al., 2002b; Maguire et al., 2000; Moore et al., 2000; Penn and Sims, 2002). Consequently, there is widespread discussion in the USA today about the most effective means to more accurately incorporate P source characteristics into risk assessment tools.

Research has clearly shown that the solubility of P in organic sources can be influenced by animal diet (Maguire et al., 2004; Toor et al., 2005), biological treatment (e.g., composting; Sharpley and Moyer, 2000), or storage practices (McGrath et al., 2005). In addition, studies have demonstrated that chemical treatment will reduce the solubility of P in biosolids (Penn and Sims, 2002) and manures (Moore et al., 2000). The environmental benefits of reducing P solubility in organic byproducts have been shown in many studies. Treating poultry litter and swine manure with Fe (e.g., FeCl3) or Al [alum, Al2(SO4)3] salts resulted in lower concentrations of WSP and soil test P in manures and manure-amended soils (Codling et al., 2000; Shreve et al., 1996) and reduced dissolved reactive phosphorus (DRP) concentrations in runoff (Moore et al., 2000; Smith et al., 2001). Similarly, soils amended with Fe- and Al-salt-treated biosolids had lower WSP and soil test P concentrations than those where lime-stabilized biosolids (with and without metal salts), biological phosphorus removal (BPR), or digested biosolids were added (Kyle and McClintock, 1995; Maguire et al., 2001). Amending soils with metal-salt-treated biosolids also resulted in lower DRP concentrations in runoff and leachate than observed with digested and BPR biosolids (Elliott et al., 2002, 2005; Penn and Sims, 2002).

The effect of source-related factors on the differential risk of P loss between organic P sources has prompted many Mid-Atlantic states to consider incorporating weighting factors, or PSCs, into regional or state P risk assessment tools to fairly account for differences in P solubility between fertilizers, manures, and biosolids. Recently, the USDA Cooperative State Research, Education, and Extension Service (CSREES) Mid-Atlantic Regional Water Group reviewed the research on this subject, including water extractions (Brandt et al., 2004; Kleinman et al., 2002b), incubation studies (Leytem et al., 2004), and rainfall simulation studies comparing runoff DRP with WSP in a wide range of biosolids and manures (Elliott et al., 2005; Kleinman et al., 2002a) and proposed the following default PSCs for DRP losses from surface-applied P sources: inorganic P fertilizer and liquid swine manure = 1.0, other untreated manures = 0.8, BPR biosolids = 0.8, alum-treated poultry litter = 0.5, and other biosolids = 0.4 (Coale et al., 2005). Some concerns have been raised, however, about simply grouping P sources into a few, discrete PSC categories, as this would require the development of standardized criteria for use in assigning manures and biosolids to PSC groups. For example, for a poultry litter to be categorized as "alum-treated," it would be necessary to specify a minimum litter Al/P ratio or guarantee that a certain amount of alum had been added to the litter, based on the number of birds grown or the size of the production facility. Alternatively, it has been proposed that research-based algorithms be developed to convert measured source WSP values into source-specific PSCs, based on the results of rainfall simulation experiments directly quantifying source effects on DRP loss (Coale et al., 2005; Elliott et al., 2006). The measured WSP/total P (TP) ratio has also been suggested as a means to rank biosolids and manures when assessing the risk of dissolved P loss (Brandt et al., 2004).

While there seems to be an emerging consensus on the value of PSCs to risk assessment for surface-applied P sources, less information is available to judge the importance of differences in organic P source solubility when manures and biosolids are incorporated into soils by tillage. Most studies have shown that tilling manures and biosolids into soils reduces soil WSP and DRP in runoff, particularly for soils with higher P sorption capacities (Daverede et al., 2003; Kleinman et al., 2002a; Penn and Sims, 2002; Tarkalson and Mikkelsen, 2004); however, the relative importance of source and soil properties on the potential for dissolved P loss from incorporated organic P sources has not been quantified in a manner that is easily adaptable to risk assessments for P loss. Consequently, our objectives in this study were to determine soil properties and P source characteristics that could (i) accurately estimate the risk of dissolved P loss to water from incorporated manures and biosolids and (ii) be easily used by farmers and nutrient management planners conducting P loss risk assessments in the Mid-Atlantic USA.


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 IMPLICATIONS AND CONCLUSIONS
 REFERENCES
 
Soil Characterization and Phosphorus Status
Working with cooperators in four Mid-Atlantic states (Delaware, Maryland, Pennsylvania, and Virginia), we identified and obtained eight agricultural soils that would typically receive land applications of organic P sources. The soil series at these sites are widely distributed in the Mid-Atlantic region and are representative of soils where organic P sources are often applied for crop production (Table 1). The Pamunkey sandy loam (fine-loamy, mixed, semiactive, thermic Ultic Hapludalf, Virginia) and Fort Mott loamy sand (loamy, siliceous, semiactive, mesic Arenic Hapludult, Delaware) are coarse-textured, well-drained soils from the Mid-Atlantic Coastal Plain. The Fallsington sandy loam (fine-loamy, mixed, active, mesic Typic Endoaquult, Delaware) and Othello silt loam (fine-silty, mixed, active, mesic Typic Endoaquult, Maryland) are also Coastal Plain soils, but are medium textured and poorly drained. The Berks clay loam (loamy-skeletal, mixed, active, mesic Typic Dystrudept, Pennsylvania) and Manor loam (coarse-loamy, micaceous, mesic Typic Dystrudept, Pennsylvania) are medium-textured, well-drained, upland soils from the Piedmont. The Davidson clay (fine, kaolinitic, thermic Rhodic Kandiudult, Virginia), also a well-drained upland Piedmont soil, is a fine-textured soil formed from ferromagnesian parent material. The Hagerstown silty clay loam (fine, mixed, semiactive, mesic Typic Hapludalf, Maryland) is a fine-textured, well-drained, limestone-derived valley soil from the Ridge and Valley physiographic province (NRCS, 2004). In general, all soils were moderately acidic with low to moderate organic matter contents (Table 1). The fine-textured Piedmont and Ridge and Valley soils generally had higher concentrations of total Fe and Al than the coarse-textured Coastal Plains soils (Table 1).


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

  		Table 1. Selected properties of the eight Mid-Atlantic soils used in the incubation study.

 
Bulk soil samples were collected at each site from the 0- to 5-cm depth, air dried at room temperature (25 ± 2°C), and sieved to pass a 2-mm screen. Soil pH (1:10 soil/deionized water) and organic matter (loss on ignition) were determined by standard methods of the University of Delaware soil-testing program (Sims and Heckendorn, 1991). Soil moisture content at field capacity was determined by the method described by Tan (1996) and particle size by the hydrometer method (Bouyoucos, 1962). Each soil was also analyzed for soil test P, Al, Fe, and Ca by Mehlich-3 extraction (M3-P, M3-Al, M3-Fe, M3-Ca; 1:10 soil/0.2 M CH3COOH + 0.25 M NH4NO3 + 0.015 M NH4F + 0.13 M HNO3 + 0.001 M EDTA), WSP (1:10 soil/deionized water, 1-h reaction time, filtration with 0.45-µm Millipore membrane), and oxalate-extractable P, Al, and Fe (Pox, Alox, Feox; 1:40 soil/0.2 M acid ammonium oxalate (pH 3), 2-h reaction time in the dark; McKeague and Day, 1966). Mehlich-3 extracts were analyzed for P, Al, Fe, and Ca and oxalate extracts were analyzed for P, Al, and Fe by inductively coupled plasma atomic emission spectroscopy (ICP-AES). Water-soluble P was analyzed colorimetrically by the molybdate blue method of Murphy and Riley (1962). The M3-PSR (Sims et al., 2002) was calculated as follows (values for P, Al, and Fe in millimoles per kilogram): M3-PSR = M3-P/(M3-Fe + M3-Al). The degree of P saturation (DPS; Schoumans, 2000) was determined as follows (values for P, Al, and Fe in millimoles per kilogram): DPS = Pox/{alpha}(Alox+Feox), where {alpha} is an empirical constant that ranges from 0.4 to 0.6; we used an {alpha} value of 0.5.

Biosolids and Manure Characterization
Ten organic P sources were selected for the incubation study including: an alum-treated poultry litter; a normal poultry litter; a liquid slurry pit dairy manure; a liquid swine manure; lime-stabilized BPR biosolids from the Little Patuxent treatment plant in Savage, MD (BPR + lime); biosolids from the Piscataway treatment plant in Accokeek, MD, that were lime stabilized before secondary treatment and received added Al for P control (lime + Al); biosolids from the Blue Plains treatment plant in Washington, DC, that were lime stabilized after secondary treatment and received FeCl3 for P control (lime + Fe); anaerobically digested biosolids from the Hampton Roads Atlantic treatment plant in Hampton Roads, VA, that received FeCl3 for struvite control; anaerobically digested biosolids from the Back River treatment plant in Baltimore, MD, that received FeCl3 addition for P control; and anaerobically digested biosolids from the Boston treatment plant in Quincy, MA. Biosolids and manures varied in physical and chemical properties but were reasonably typical in composition for these types of organic byproducts. Solids content ranged from <1% for the liquid swine manure to >76% for poultry litters. Organic P source pH was near neutral (<7.7) for digested biosolids, Fe-treated biosolids, dairy manure, and swine manure, slightly alkaline (8.1–8.6) for alum-treated and normal poultry litters, and alkaline (>10.3) for lime-stabilized biosolids. All organic P sources were refrigerated at 4°C on receipt and analyzed "as-is" for TP, Al, Fe, and Ca by microwave-assisted HNO3 digestion (USEPA, 1986) and WSP (1:200 dry manure/deionized water, 1-h reaction time, filtration through Whatman no. 40 filter paper; Wolf et al., 2005).

Incubation Study
Each of the 10 organic P sources was incorporated (in duplicate) into 250 g of each of the eight soils at a rate of 60 mg TP kg–1 (approximately 135 kg P ha–1), which is representative of an average P application rate when organic P sources are applied to meet crop N needs. An inorganic P source (KH2PO4) was applied at the same rate to represent commercial P fertilizer and an unamended soil was used for each soil. This total P rate resulted in the addition of 1.1 to 6.3 kg WSP ha–1 for biosolids, and 21, 36, 107, and 116 kg WSP ha–1 for alum-treated poultry litter, normal poultry litter, dairy manure, and swine manure, respectively. During P source incorporation, soils were brought to 80% field capacity (Tan, 1996) with deionized water. After incorporation, the soils were transferred to 240-mL polyethylene cups with two holes in the snap cap to prevent anaerobic conditions and allow gas exchange during incubation. Soils were incubated at room temperature for 180 d. Soil moisture content was maintained at 80% field capacity by adding deionized water to the sample cups weekly (based on sample weight). Subsamples were taken at 2, 30, and 180 d (82.0 ± 0.5, 100 ± 1, and 2.000 ± 0.005 g, respectively) and analyzed immediately for WSP by the molybdate blue method of Murphy and Riley (1962). Subsamples collected at 30 d were air dried, ground, and analyzed for pH, M3-P, M3-Ca, M3-Al, M3-Fe, and M3-PSR, as described above.

Statistical Analysis
Statistical analysis of the incubation data was performed using the PROC MIXED procedure (SAS Institute, Cary, NC). The effect of P source on soil P was evaluated for each individual soil due to a positive soil x P source interaction for all measured variables. Significant differences between P sources for each soil were evaluated using the Tukey's Studentized range test at a probability level of 0.05. Linear relationships between the initial properties of soils or organic P sources and soil WSP after incubation were analyzed using PROC CORR. We also used the PROC REG procedure (RSQUARE option) to identify the soil and organic P source properties that could best predict soil WSP at each sample date.


   RESULTS AND DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
IMPLICATIONS AND CONCLUSIONS
 REFERENCES
 
Soil Characterization and Phosphorus Status
For the eight soils collected for the incubation study, M3-P ranged from 27 to 141 mg kg–1 and was below the values used in the Mid-Atlantic region to restrict land application of organic P sources (M3-P = 150–200 mg kg–1; Sims et al., 2002; Beegle et al., 1998). Assuming no limiting site characteristics (e.g., erosion, runoff, proximity to surface water) and reasonable P source management practices, land application of manures and biosolids would currently be permitted for all eight soils. We also measured TP and WSP in these soils, which ranged from 98 to 1155 and 0.6 to 7.4 mg kg–1, respectively. Measures of soil P saturation ranged from 0.02 to 0.16 for M3-PSR and from 20 to 47% for DPS (Table 1). Total P was poorly correlated with M3-P (plant available) and WSP (r = 0.17 and –0.19, not significant at P = 0.05), but soil WSP was highly correlated with M3-P, M3-PSR, and DPS (r = 0.78, 0.93, and 0.94, respectively, significant at P = 0.001).

Biosolids and Manure Characterization
Total P concentrations in manures and biosolids ranged from 12.6 to 56.9 g kg–1 and were similar to values reported in other studies (Table 2; Kleinman et al., 2005; Stehouwer et al., 2000). Biosolids, especially those where metal salts [e.g., Al2(SO4)3, FeCl3] were added during wastewater treatment, had much lower WSP values (<5% of TP) than untreated manures (27–88% of TP; Table 2). As shown in past research, treating broiler litter with alum in poultry production houses also reduced WSP (Sims and Luka-McCafferty, 2002). Since the concentrations of TP and WSP in organic P sources varied widely, we compared biosolids and manures based on the proportion of TP that is soluble (e.g., WSP/TP ratio). Organic P source WSP/TP values were similar to those reported in several studies (Brandt et al., 2004; Kleinman et al., 2002a, 2005; Sharpley and Moyer, 2000; Withers et al., 2001), with biosolids having substantially lower values than untreated manures (Table 2). In addition, the WSP/TP values were well correlated with default PSC values (r = 0.92, P = 0.001), which were developed for the PSI to account for differences in P solubility of organic P sources (Coale et al., 2005; Leytem et al., 2004).


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

  		Table 2. Selected properties of the 10 organic P sources used in the incubation study.

 
Consistent with past research (Brandt et al., 2004; Leytem et al., 2004), when a diverse group of organic P sources was considered, TP was a poor predictor of WSP concentrations in manures and biosolids (r2 = 0.67 for all sources, but r2 < 0.001 when swine manure, clearly an outlier for this correlation, was excluded; Fig. 1a ). This provides further evidence that risk assessment tools such as the PSI should account for differences in P solubility between organic P sources and not rely solely on the total amount of P applied. Short-term P solubility, and thus the relative risk of dissolved P loss from surface applications and the release of P from solid phases in biosolids and manures on incorporation into soils, could be much better predicted from total Al + Fe + Ca (mol kg–1) than TP (r2 = 0.74, P = 0.001, without swine manure; r2 = 0.57, P = 0.001, with swine manure). Our findings were very consistent with past studies by Elliott et al. (2005) and Penn and Sims (2002), who also reported curvilinear relationships between WSP and the molar concentrations of metals (Al, Fe, and Ca) in biosolids and manures (Fig. 1b).


Figure 1
View larger version (19K):
[in this window]
[in a new window]

  		Fig. 1. Relationship between organic P source (a) water-soluble P (WSP) and total P and (b) WSP and total Al + Ca + Fe; SM = liquid swine manure.

 
Analysis of these biosolids and manures using x-ray absorption near edge structure (XANES) spectroscopy identified PO4 sorbed to Al hydroxide, hydroxylapatite [Ca5H(PO4)3•2.5H2O], and phytic acid as the dominant P species in the dairy manure, normal and alum-treated poultry litters, BPR + lime and lime + Al biosolids; PO4 sorbed to ferrihydrite, hydroxylapatite, and phytic acid in lime + Fe biosolids; and PO4 sorbed to ferrihydrite, hydroxylapatite, ß-tricalcium phosphate [ß-Ca3(PO4)2], and in many cases PO4 sorbed to Al hydroxide in Fe-treated and digested biosolids (Shober et al., 2006). There is also evidence (e.g., XANES analysis, organic P source WSP/TP and total Al/P ratios) that a portion of the TP in dairy manure and normal poultry litter that was identified as Al hydroxide sorbed P could be a soluble hydrated salt (e.g., K or Na phosphate) or weakly bound P (e.g., outer sphere surface complexes; Shober et al., 2006; Toor et al., 2005). These results suggest that the lower P solubility in biosolids and alum-treated poultry litter than in untreated manures was due to the precipitation of Al or Fe hydroxide and sorption of PO4 or coprecipitation of phosphate with Al or Fe hydroxides (Table 2). Results also suggested that the addition of lime during wastewater treatment reduced WSP by promoting the precipitation of hydroxylapatite in lime-stabilized biosolids.

Effects of Manures and Biosolids on Soil Phosphorus Mehlich 3 Phosphorus and Phosphorus Saturation Ratio
Despite the wide range in P solubility of the 10 P sources used in this study, we noted very little difference in source effects on M3-P and soil P saturation for amended soils (data not shown). The average effect, for all eight soils and 10 P sources, of adding 60 mg TP kg–1 soil was an increase in M3-P of 25 ± 5 mg kg–1 and M3-PSR of 0.03 ± 0.01. Furthermore, there was no appreciable difference in the increase in M3-P and M3-PSR from untreated manures and inorganic P (mean of eight soils for swine manure, dairy manure, normal poultry litter, and KH2PO4: 26 ± 9 mg kg–1 and 0.03 ± 0.01) compared with alum-treated poultry litter and biosolids (mean: 24 ± 8 mg kg–1 and 0.03 ± 0.01). There was a clear trend, however, for increases in M3-P and M3-PSR to be smaller for finer textured soils that had higher Al and Fe contents than for the coarse-textured soils with low Al and Fe content. For example, average changes (across all P sources) in M3-P and M3-PSR were negatively correlated with clay content (r2 = –0.72, P = 0.01, for M3-P; r2 = –0.81, P = 0.05, for M3-PSR) and soil total Al + Fe (moles per kilogram; r2 = –0.81, P = 0.05, for M3-P; r2 = –0.87, P = 0.01, for M3-PSR). These results suggested that the risk for dissolved P losses would be lower when organic P sources are incorporated into finer textured soils with high Al and Fe concentrations than for coarser textured soils with low Al and Fe. Studies have shown, however, that fine-textured soils have a higher risk of erosion losses and potential for substantial particulate P losses (Uusitalo et al., 2001).

The addition of organic P sources also affected M3-Al, -Ca, and -Fe, which can have implications for long-term solubility of P in amended soils. The most notable change, following the incorporation of dairy manure and lime-stabilized biosolids, was an increase in M3-Ca ranging from 35 to 111% relative to the unamended soil (1215, 1164, 1413, 1749 mg kg–1 for dairy manure, BPR + lime, lime + Al, and lime + Fe biosolids, vs. 909 mg kg–1 for the unamended soil). The dairy manure and lime-stabilized biosolids contained high concentrations of hydroxylapatite, which was very insoluble at the pH (7.7–12.3) of these organic P sources; however, when these materials were incorporated into acidic soils, the pH of the amended soils (5.3–7.3) was much lower than the pH of the organic P source. This may have solubilized some of the hydroxylapatite, resulting in an increase in soil M3-Ca. Trends also indicated that the incorporation of Fe-treated biosolids increased M3-Fe, by an average of 17% (226, 236, 225, 220 for lime + Fe biosolids, Hampton Roads and Back River Fe-treated biosolids, and digested biosolids, vs. 202 mg kg–1 for the unamended soil), suggesting, as was reported by Maguire et al. (2001), that additions of Fe-stabilized biosolids could also add reactive Fe oxides to soils. With time, added Fe oxides have the potential to increase soil P sorption capacity, thereby decreasing the risks of dissolved P loss by runoff and leaching.

Water-Soluble Phosphorus
We assessed the potential environmental impacts of the differential solubility of manures and biosolids on water quality by measuring soil WSP, which has been shown in many studies to be well correlated with dissolved P losses in surface runoff and leachate (Maguire and Sims, 2002; Pote et al., 1999; Westermann et al., 2001). Water-soluble P concentrations in the amended soils generally paralleled the solubility of the organic P sources added, but were also influenced markedly by soil properties and particularly by soil P saturation (Fig. 2 ). Average soil WSP values (across all eight soils) after 2 d of incubation, a time period normally adequate for fast sorption reactions to occur with added soluble P, were highest with the inorganic P source and the untreated animal manures (swine manure > fertilizer > dairy manure > normal poultry litter = 8.0, 6.3, 6.3, and 5.2 mg kg–1, respectively) and lowest with the alum-treated poultry litter (2.9 mg kg–1) and biosolids (3.4 ± 0.7 mg kg–1). Average WSP values for the unamended soils were 2.7 mg kg–1. Source effects on soil WSP were most pronounced for soils with higher degrees of P saturation (M3-PSR, DPS), such as the Berks and Pamunkey soils, which were near or above the environmental P saturation values identified by Sims et al. (2002; M3-PSR > 0.15, DPS > 40%). Even in these two soils, however, biosolids and alum-treated poultry litter did not significantly increase soil WSP concentrations relative to the unamended soil. Conversely, in soils with lower P saturation values (M3-PSR ≤ 0.07, DPS ≤ 26%; Table 1), such as the Fort Mott, Manor, and especially the Davidson soil, very few significant differences in soil WSP were found between inorganic P, untreated manures, alum-treated poultry litter, and lime-stabilized, digested, or Fe-treated biosolids (Fig. 2).


Figure 2
View larger version (38K):
[in this window]
[in a new window]

  		Fig. 2. Effect of incorporation of organic P sources and KH2PO4 at 135 kg P ha–1 on soil water-soluble P in a wide range of Mid-Atlantic soils at the 2-d sampling date. Values within the same soil with the same letter are not significantly different at P < 0.05 using Tukey's Studentized range test. M3-PSR = Mehlich-3 phosphorus saturation ratio; APL = alum-treated poultry litter; B1 = biological P removal + lime biosolids; B2 = lime + Al biosolids; B3 = lime + Fe biosolids; B4 = Fe-treated biosolids (Hampton Roads); B5 = Fe-treated biosolids (Back River); B6 = anaerobically digested biosolids; DM = liquid slurry pit dairy manure; F = fertilizer (KH2PO4); NPL = normal poultry litter; SM = liquid swine manure; U = unamended soil.

 
We also observed that soil WSP concentrations at 2 d were well correlated with those measured at 30 d (r2 = 0.79, P = 0.001) and 180 d (r2 = 0.70, P = 0.001; Fig. 3 ). At the 30-d sampling date, soil WSP remained the same or increased slightly (median change = 0.23 mg kg–1) relative to the 2-d samples for most P sources; however, consistent decreases in soil WSP were noted with dairy manure and the lime + Fe and digested biosolids. After 180 d of incubation, soil WSP values were lower for normal poultry litter, swine manure, dairy manure, and fertilizer than the 2- and 30-d samples, reflecting slow sorption reactions that occur following P additions to soils, usually attributed to slow diffusion of phosphate into micropores followed by sorption at colloidal surfaces (Barrow, 1983; Torrent et al., 1992; van Riemsdijk et al., 1984). Differences in soil WSP between all P sources were considerably smaller by 180 d compared with values reported after incubation for 2 d, but the original trend noted at 2 d persisted. For example, mean soil WSP values (180 d), across all soils, were 4.2, 3.5, 3.5, and 3.4 mg kg–1 for swine manure, fertilizer, dairy manure, and normal poultry litter, respectively, compared with 2.6 mg kg–1 for alum-treated poultry litter, 2.8 ± 0.4 mg kg–1 for the biosolids, and 2.2 mg kg–1 for unamended soils. The influence of soil P saturation on soil WSP also remained evident throughout the study, as seen in Fig. 4 , which illustrates the relationship between initial M3-PSR values for the eight soils and soil WSP in soils amended with fertilizer, swine manure, and Back River Fe-treated biosolids at the 2- and 180-d sampling dates. Similar relationships were observed with the other manures and biosolids for soil WSP and M3-PSR (data not shown).


Figure 3
View larger version (23K):
[in this window]
[in a new window]

  		Fig. 3. Comparison of soil water-soluble phosphorus (WSP) concentrations after incubation of organic P sources and Mid-Atlantic soils for 2 d with soil WSP concentrations after incubation for (a) 30 d and (b) 180 d. APL = alum-treated poultry litter; B1 = biological P removal + lime biosolids; B2 = lime + Al biosolids; B3 = lime + Fe biosolids; B4 = Fe-treated biosolids (Hampton Roads); B5 = Fe-treated biosolids (Back River); B6 = anaerobically digested biosolids; DM = liquid slurry pit dairy manure; F = fertilizer (KH2PO4); NPL = normal poultry litter; SM = liquid swine manure; U = unamended soil.

 

Figure 4
View larger version (18K):
[in this window]
[in a new window]

  		Fig. 4. Relationship between initial soil Mehlich-3 phosphorus saturation ratio (M3-PSR) of the eight Mid-Atlantic incubation soils and soil water-soluble phosphorus (WSP) after incubation for (a) 2 d and (b) 180 d following the incorporation of KH2PO4 (F), liquid swine manure (SM), and Fe-treated biosolids (B5) at 135 kg total P ha–1.

 
Based on the trends in soil WSP following incorporation of organic P sources, we can draw two main observations. First, regardless of the type of P source added, soil WSP increased linearly with the degree of soil P saturation at both the 2- and 180-d sampling dates. This is consistent with the results of Leytem and Sims (2005), who reported that the slope between soil WSP and the amount of broiler litter P added increased with the degree of soil P saturation. Second, the extent of the increase in soil WSP, as indicated by the slopes in Fig. 4, was initially greater with the more-soluble P sources (e.g., swine manure, dairy manure, and normal poultry litter) than the alum-treated poultry litter and Back River Fe-treated biosolids. Thus, the short-term risk of dissolved P loss when organic P sources are incorporated into soils will be affected by the solubility of the source and soil P saturation. With time, however, differences between slopes diminished and source effects become less important than soil P saturation. We also found that slightly better predictions of soil WSP could be obtained if P saturation was measured by oxalate extraction (DPS; r2 = 0.71–0.90 at 2 d and 0.44–0.85 at 180 d) rather than M3-PSR (data not shown).

Our soil WSP data provide preliminary support for the development of PSC values that can rank the relative risk of incorporated organic P sources, in addition to those that are surface applied, on dissolved P losses. They also suggest the potential need to incorporate a measure of soil P saturation (e.g., M3-PSR or DPS) into PSIs and other P risk assessment tools as a rapid measure of the effect of different soils on soil WSP after incorporation of manures and biosolids, as recommended for the Florida PSI by Nair et al. (2004). Finally, our results also suggest that a 2-d incubation of organic P sources with soils could be a rapid, inexpensive laboratory test to predict, with reasonable accuracy, longer term effects of a wide variety of soil–source combinations on soil P solubility.

Predicting Changes in Soil Water-Soluble Phosphorus with Source and Soil Properties
Rapid, inexpensive, and widely available tests that could predict how organic P sources and soil properties will affect soil WSP concentrations would improve our ability to assess the risk of dissolved P loss from soils to surface and groundwaters. Results of our incubation study showed, for a wide range of Mid-Atlantic soils amended with a diverse set of P sources (manures, biosolids, or inorganic P, excluding the unamended soil data), that soil WSP concentrations at 2, 30, and 180 d were better correlated with measures of initial soil P saturation (DPS, r = 0.69, 0.77, and 0.79; and M3-PSR, r = 0.65, 0.74, and 0.74, all at P = 0.001) than with M3-P (r = 0.54, 0.62, and 0.57, all at P = 0.001). Phosphorus source properties alone were not as well correlated with soil WSP as measures of soil P saturation and, unlike M3-PSR and DPS, source-related correlations decreased with incubation time. The best manure and biosolids properties for prediction of soil WSP after incubation for 2, 30, and 180 d were source WSP (r = 0.41, 0.38, and 0.21, respectively, all at P = 0.001), WSP/TP (r = 0.49 at P = 0.001, 0.34 at P = 0.001, and 0.21 at P = 0.01, respectively) and PSCs (r = 0.49, 0.38, and 0.23, respectively, all at P = 0.001).

Multiple regression equations including both source and soil properties were more effective at predicting soil WSP, and the most accurate predictions could be achieved (up to r2 values of 0.81 at P = 0.001) with multiple regression models using four or more parameters; however, this required inclusion of soil and P source properties that are more difficult and time consuming to measure. Two examples illustrate this point for WSP at the 2-, 30-, and 180-d sampling dates. By using soil DPS and source WSP, we obtained r2 values of 0.64 (P = 0.001), 0.73 (P = 0.001), and 0.66 (P = 0.001). Adding the initial value of WSP in unamended soils improved r2 values to 0.68 (P = 0.001), 0.77 (P = 0.001), and 0.76 (P = 0.01). The addition of total Ca in the P source to the regression equation as a fourth parameter further improved r2 values to 0.78 (P = 0.001), 0.80 (P = 0.001), and 0.78 (P = 0.001). Currently, soil and P source WSP tests are not offered as routine tests by commercial testing labs in the Mid-Atlantic region; total Ca is offered as a manure test at an additional cost. This points out one of the challenges with the development of practical risk assessment tools for P. While an improved ability to predict soil WSP by testing soils, manures, and biosolids is clearly desirable, it is only of value if the tests can be conducted widely and results disseminated to end users such as county Extension personnel, farmers, and nutrient management consultants. This is unlikely to occur if more complex tests (oxalate DPS) or additional tests (total Al, Fe, and Ca in P sources, initial soil WSP) are required. Note that some advancement has been made in providing soil P saturation information on a routine basis. For example, the University of Delaware soil testing program added M3-PSR to its routine soil test in 2002 as a measure of soil P saturation, simply by using ICP-AES to measure Al and Fe, as well as P, in M3 extracts and has since analyzed thousands of soil samples for M3-PSR. The M3-PSR has been shown in several studies to be well correlated to soil DPS (r2 = 0.92, P = 0.001; Sims et al., 2002) and to be an acceptable means to predict P loss by runoff and leaching (Breesuwsma et al., 1995; Khiari et al., 2000; Kleinman and Sharpley, 2002; Sims et al., 2002). In our study, the ability to predict soil WSP after incubation for 2, 30, and 180 d was only slightly less when M3-PSR was used instead of DPS as the measure of soil P saturation in multiple regression equations with source WSP (r2 = 0.59, 0.68, and 0.59 for 2, 30, and 180 d, respectively vs. r2 = 0.64, 0.73, and 0.66, respectively, all at P = 0.001).

While the P source WSP method (Wolf et al., 2005) used in our study is not currently offered as a routine test, it has been used to determine soluble P on manure, biosolids, and other residual samples for research purposes (Ann Wolf, personal communication, 2006). The P source WSP test has been shown to be a reliable means to predict DRP concentrations in runoff from surface-applied manures and biosolids (Elliott et al., 2005; Kleinman et al., 2002a; Vadas et al., 2005). Additionally, this method could be easily incorporated as a routine test by most commercial testing laboratories. Commercial availability of the source WSP test would also allow calculation of the P source WSP/TP ratio. The WSP/TP ratio represents the fraction of total P that is water soluble and allows a more direct comparison of environmentally relevant P in biosolids and manures with varying chemical and physical properties (Brandt et al., 2004). Previous research has shown a strong correlation (r > 0.79, P = 0.001) between P source WSP/TP ratio and DRP in runoff when 14 manure and biosolids were broadcast on the soil surface (Sullivan et al., 2005). Our results showed that a multiple regression equation that included initial soil M3-PSR and source WSP/TP was slightly better at predicting soil WSP(r2 = 0.68 at P = 0.001, 0.66 at P = 0.01, and 0.61 at P = 0.001 at 2, 30, and 180 d, respectively) than M3-PSR and source WSP. On the other hand, no additional organic P source testing would be required to use default PSC values as a predictor of soil WSP following organic P source incorporation. When default PSCs were used in multiple regression analysis with soil M3-PSR, r2 values for the prediction of soil WSP at 2, 30, and 180 d were 0.67, 0.69, and 0.60 (all at P = 0.001), respectively. Our data show that combining source default PSC and M3-PSR measurements results in a simple, inexpensive means to estimate soil WSP, and thus the potential for DRP losses in runoff, for incorporated organic P sources. The ability of soil P saturation and default PSCs to predict soil WSP suggests that ongoing work to develop a standard organic P source WSP test to allow calculation of source-specific PSCs, as suggested by Coale et al. (2005) and Elliott et al. (2006), may be unnecessary. It is important to note, however, that our study only evaluated 10 organic P sources and it may be beneficial to use source-specific PSCs (based on WSP) to deal with variations in physical and chemical properties (e.g., variable amounts of chemical salts, bedding materials, compost materials, etc.) that occur across a wide number of organic P sources.


   IMPLICATIONS AND CONCLUSIONS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 IMPLICATIONS AND CONCLUSIONS
REFERENCES
 
Six states (Florida, Georgia, Maryland, Pennsylvania, Tennessee, and Virginia) have now modified their PSIs to account for differences in P solubility between sources by incorporating weighting factors, such as PSCs, into their P risk assessment tools. While the use of PSCs has been shown to be an effective way to assess the relative risk of dissolved P transport for surface applications (Elliott et al., 2005), our results clearly show that PSCs alone will not accurately predict dissolved P losses when fertilizers and organic sources are incorporated into soils that differ in P saturation. Therefore, we propose adding a simple numerical rating factor (PSC–PSR factor) that integrates soil P saturation and PSCs to P loss risk assessment tools (Table 3).


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

  		Table 3.  Proposed approach to combine default source P source coefficients{dagger} (PSCs) and soil P saturation values to rank the risk of dissolved P loss when organic P sources are incorporated into soils.

 
As shown in Table 3, we suggest developing a matrix of PSC–PSR risk categories by combining five soil P saturation categories and four default PSC values (Table 2). The soil P saturation categories are based on the numerical relationship between DPS and runoff DRP reported by Vadas et al. (2005) in a review of a wide range of research investigating the influence of soil P saturation on DRP in runoff. Note that we also include in Table 3 the M3-PSR values that correspond to the DPS categories, based on the relationship reported by Sims et al. (2002) for 465 soils from the Mid-Atlantic USA (M3-PSR = 0.0042 x DPS – 0.019, r2 = 0.92, P = 0.001). Since the PSC–PSR factor is designed to take the place of the PSC in cases where organic P sources are incorporated into the soil, it must range in value from zero to one, but have a value less than the PSC alone to account for the reduced risk of P loss that accompanies organic P source incorporation. Therefore, we calculated PSC–PSR risk categories for each soil and P source combination used in our incubation study by dividing default PSC values by the soil P saturation rating factors (Table 3). For the soils and P sources used in our study, calculated PSC–PSR factors ranged from 0.08 (e.g., Davidson soil, biosolids) to 0.50 (Pamunkey soil, swine manure). Calculated PSC–PSR risk values were able to accurately predict soil WSP at 2, 30, and 180 d (r2 = 0.96, 0.85, and 0.80, all at P = 0.001, respectively) for a diverse group of P sources and soils common to the Mid-Atlantic region (Fig. 5 ).


Figure 5
View larger version (18K):
[in this window]
[in a new window]

  		Fig. 5.   Prediction of soil water-soluble phosphorus (WSP) after incorporation of organic P sources and KH2PO4 into eight Mid-Atlantic soils using a Mehlich-3 phosphorus saturation ratio (M3-PSR)–phosphorus source coefficient (PSC) rating system.

 
The PSC–PSR factor would improve PSIs (without the need for additional soil or organic P source testing) by accounting for the risk of dissolved P loss to water when fertilizers, manures, or biosolids are surface applied (multiply total P loading rate by PSCs only) or incorporated into the soil (multiply total P loading rate by PSC-PSR factor). Additionally, inclusion of the PSR-PSC factor in the PSI should allow further simplification of the risk assessment tools by elimination of application methods and timing factors, provided an appropriate time frame for incorporation of organic P sources is established (e.g., 5 d–1 wk). Before default PSCs and the PSC-PSR factor could be accepted for use in P risk assessment tools, however, it will be necessary to develop standardized criteria for organic P sources (e.g., the amount of alum or Al/P ratio required to categorize a poultry litter as alum-treated). In addition, it would be beneficial to examine the relationship between soil WSP (or runoff DRP) and the PSC–PSR factors with a broader range of organic P sources and at variable total P rates.


   ACKNOWLEDGMENTS
 
The Metropolitan Washington Council of Governments (MWCOG) and the Institute of Soil and Environmental Quality (ISEQ) at the University of Delaware provided funding for this research. Special thanks are extended to Dr. Peter Kleinman (USDA-ARS), Dr. Frank Coale (University of Maryland), Dr. Greg Evanylo (Virginia Tech), Ms. Diane Shields (Delaware NRCS), and their support staff for assistance locating and collecting the soils used in this study. We would also like to thank Karl Berger at MWCOG for his assistance in obtaining biosolids samples and for providing insights into factors affecting land application of biosolids in the Mid-Atlantic USA.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 IMPLICATIONS AND CONCLUSIONS
 REFERENCES
 
All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permission for printing and for reprinting the material contained herein has been obtained by the publisher.

Received for publication July 7, 2006.


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




This article has been cited by other articles:


Home page
 J. Environ. Qual.Home page
M. C. Smith, J. W. White, and F. J. Coale
Evaluation of Phosphorus Source Coefficients as Predictors of Runoff Phosphorus Concentrations
J. Environ. Qual., February 6, 2009; 38(2): 587 - 597.
[Abstract] [Full Text] [PDF]

Home page
 J. Environ. Qual.Home page
A. L. Shober and J. T. Sims
Evaluating Phosphorus Release from Biosolids and Manure-Amended Soils under Anoxic Conditions
J. Environ. Qual., January 13, 2009; 38(1): 309 - 318.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Shober, A. L.
Right arrow Articles by Sims, J. T.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Shober, A. L.
Right arrow Articles by Sims, J. T.
Agricola
Right arrow Articles by Shober, A. L.
Right arrow Articles by Sims, J. T.
Related Collections
Right arrow Phosphorus
Right arrow Nutrient Management

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