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


     

Published online 1 May 2008
Published in J Environ Qual 37:1180-1189 (2008)
DOI: 10.2134/jeq2007.0535
© 2008 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 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 Google Scholar
Google Scholar
Right arrow Articles by Agyin-Birikorang, S.
Right arrow Articles by Brinton, S. R.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Agyin-Birikorang, S.
Right arrow Articles by Brinton, S. R.
Agricola
Right arrow Articles by Agyin-Birikorang, S.
Right arrow Articles by Brinton, S. R.
Related Collections
Right arrow Phosphorus
Right arrow Residue management
Right arrow Soil Chemistry
Right arrow Other Soil Management
Right arrow Municipal Waste

TECHNICAL REPORTS

Evaluating Phosphorus Loss from a Florida Spodosol as Affected by Phosphorus-Source Application Methods

S. Agyin-Birikoranga,*, G. A. O'Connora and S. R. Brintonb

a Soil and Water Sci. Dep., Univ. of Florida, 106 Newell Hall, P.O. Box 110510, Gainesville, FL 32611-0510
b Crop Science Dep., North Carolina State Univ., 4104 Williams Hall, P.O. Box 7620, Raleigh, NC 27695-7620

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

Received for publication October 8, 2007.

   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 Materials and Methods
 Results and Discussion
 Summary and Conclusions
 REFERENCES
 
Incorporating applied phosphorus (P) sources can reduce P runoff losses and is a recommended best management practice. However, in soils with low P retention capacities, leaching can be a major mechanism for off-site P loss, and the P-source application method (surface or incorporation) may not significantly affect the total amount of off-site P loss. We utilized simulated rainfall protocols to investigate effects of P-source characteristics and application methods on the forms and amounts of P losses from six P sources, including five biosolids materials produced and/or marketed in Florida, and one inorganic fertilizer (triple superphosphate). A typical Florida Spodosol (Immokalee fine sand; sandy, siliceous, hyperthermic Arenic Alaquods) was used for the study, to which the P sources were each applied at a rate of 224 kg P ha–1 (approximately the P rate associated with N-based biosolids applications). The P sources were either surface-applied to the soil or incorporated into the soil to a depth of 5 cm. Amended soils were subjected to three simulated rainfall events, at 1-d intervals. Runoff and leachate were collected after each rainfall event and analyzed for P losses in the form of soluble reactive P (SRP), total dissolved P (TDP), total P (TP), and bioavailable P (BAP) (in runoff only). Cumulative masses (runoff + leachate for the three rainfall events) of P losses from all the P sources were similar, whether the amendments were surface-applied or incorporated into the soil. The solubility of the amendment, rather than application method, largely determines the P loss potential in poorly P-sorbing Florida Spodosols.

Abbreviations: Alox, oxalate extractable aluminum • ANOVA, analysis of variance • BAP, bioavailable phosphorus • BPR, biological phosphorus removal • DDI, distilled, deionized water • DPS, degree of phosphorus saturation • Feox, oxalate extractable iron • GRU, Gainesville regional utilities • ICP–AES, inductively coupled plasma–atomic emissions spectroscopy • OCUD, Orange County Utilities Division • PP, particulate phosphorus • PSC, phosphorus source coefficient • PSI, phosphorus saturation index • PWEP, percentage of water-extractable phosphorus • SRP, soluble reactive phosphorus • TDP, total dissolved phosphorus • TP, total phosphorus • TSP, triple superphosphate. WEP, water-extractable phosphorus


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
Materials and Methods
 Results and Discussion
 Summary and Conclusions
 REFERENCES
 
MOST states in the United States are developing site assessment methods to identify critical source areas for P loss from agricultural watersheds to address P nutrient management (Weld et al., 2000). According to Sharpley et al. (2003), 47 states have adopted a P index approach. Site P-indices evaluate the relative risk of P loss based on site characteristics influencing P transport, the type of P-source applied, and various crop and management practices (Lemunyon and Gilbert, 1993; Graetz et al., 2004; Leytem et al., 2004; Elliott et al., 2005; Elliott and O'Connor, 2007).

In addition to accounting for the differences in solubility of P sources in the various P indices, accounting for P loss mechanisms from receiving soils is equally important in potential P loss assessment. Unlike N, leaching of P has not traditionally been viewed as a major ground water problem. However, downward movement of P from organic wastes is potentially significant in areas with coarse-textured soils of low P-sorbing capacity and shallow ground water (Eghball et al., 1996; Harris et al., 1996; Lu and O'Connor, 2001; O'Connor et al., 2005). Surface water and ground water systems are often hydrologically linked, and leached P can move to surface water via lateral subsurface flow (He et al., 1999). Deep leaching of P through sandy soils of the Atlantic Coastal Plain is a particular concern in states with areas of intensive animal agriculture (Sims et al., 1998; Novak et al., 2000). Such conditions present unique challenges for P management.

The P indices for many states, including Florida, award credits for runoff control practices (strip-cropping, mulching, terracing, etc.) known to minimize overland P transport. In the Florida P index (Graetz et al., 2004), surface application of P sources is assigned a high vulnerability rating, compared with soil incorporation of the P sources. For example, when solids (fertilizers, manures, compost, biosolids) are incorporated within 1 d of application, the potential for P loss is assumed to be low (site vulnerability value = two). Solids surface-applied, and not incorporated into the soil, are assumed to have high potential for P loss, particularly through surface runoff (site vulnerability value = six). Thus, incorporation of soil amendments is encouraged in an attempt to reduce surface runoff of applied P sources and to reduce a site's vulnerability rating.

Most studies, particularly rainfall simulation experiments to assess edge-of-field P loss from applied P sources, were conducted with surface-applied P sources. Few studies (Penn and Sims, 2002; O'Connor and Elliott, 2006) have examined P losses from soils when the P sources were incorporated into the soil. O'Connor and Elliott (2006) hypothesized that if a P-source is mixed in and equilibrated with the soil, sorption of P by soil surfaces would lessen the amount of P susceptible to runoff. The authors showed that when Largo biosolids [a biological phosphorus removal (BPR) material with high water soluble P content] was incorporated, rather than surface-applied, runoff dissolved P was reduced by ~10-fold (O'Connor and Elliott, 2006). In contrast, Penn and Sims (2002) showed that even when various biosolids types were mixed thoroughly with soil before rainfall simulation, the amounts of dissolved P lost in runoff was large. Researchers in both studies assumed P leaching was negligible due to the high P sorption capacity of the soils used, and leached P was not quantified. In coarse-textured Florida Spodosols, however, P leaching is the dominant P loss mechanism (Elliott et al., 2002; O'Connor et al., 2005; Alleoni et al., 2008). We hypothesized that P losses in runoff and leachate from P sources depend on the P solubility of the material, and that the amounts of P lost from Florida Spodosols will be similar whether P sources were surface-applied or incorporated into the soil. The objectives of the study were, therefore, to (i) determine runoff and leachate P losses from a typical Florida Spodosol amended with P sources of different solubilities, and (ii) evaluate effects of P-source application method on the forms and amounts of P loss from the applied P sources.


   Materials and Methods
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Materials and Methods
Results and Discussion
 Summary and Conclusions
 REFERENCES
 
Soil and Phosphorus Sources Used
A typical Florida Spodosol (Immokalee fine sand, sandy, siliceous, hyperthermic Arenic Alaquods) was utilized for the study. Native Immokalee sand, not contaminated by manure depositions and having "very low" soil test P, was collected from the University of Florida Research and Education Center in Immokalee, FL. Multiple random bulk samples were collected from the A horizon (0–15 cm) and thoroughly mixed to yield a composite sample. Some chemical characteristics of the soil used are presented in Table 1 .


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

  		Table 1. Selected physicochemical properties of the Immokalee fine sand used for the study. Numbers are mean values of six replicates ± 1 SD (Agyin-Birikorang and O'Connor, 2007).

 
Five biosolids (Lakeland, Orange County, Gainesville, Disney, and Milorganite biosolids), and one mineral fertilizer [triple superphosphate (TSP)] were used in this study. Selected physicochemical characteristics of the biosolids and the TSP are presented in Table 2 . The Lakeland biosolids, generated by autothermal thermophilic aerobic digestion, was obtained from Glendale Water Reclamation Facility, Lakeland, FL, and possess physicochemical characteristics similar to a biologically P removal (BPR) biosolids material. The Orange County biosolids, an anaerobically digested (BPR) material was obtained from Orange County Utilities Division, Orlando, FL. The Disney material (Orlando, FL) was produced by composting a mixture of dewatered primary and secondary wastewater solids with wood chips via the aerated static pile method, and stabilized with Al and Fe to reduce the labile P content of the end product. The Gainesville biosolids was obtained from the water reclamation facilities of the Gainesville Regional Utilities (Gainesville, FL) and was produced through aerobic digestion. The Gainesville biosolids was a BPR-like biosolids material. Milorganite biosolids, obtained from Milwaukee Metropolitan Sewerage District, Milwaukee, WI, was generated from anaerobically digested material that was heat-dried and pelletized. The Milorganite biosolids was stabilized with Fe salts to decrease P solubility of the biosolids.


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

  		Table 2. Selected chemical properties of the P sources used for the study. Numbers are mean values of six replicates ± 1 SD.

 
Characterization of Phosphorus Sources
Biosolids and the mineral fertilizer (TSP) were analyzed for percentage total solids, water-extractable P (WEP), and total P (TP) concentrations. Percentage solids were determined by drying materials at 105°C (Gardner, 1986), and pH measurements were performed on the materials (Thomas, 1996). Total C and N concentrations were determined in the biosolids by combustion at 1010°C using a Carlo Erba analyzer (NA-1500 CNS, Carlo Erba, Milan, Italy) as outlined in Nelson and Sommers (1996). For WEP analysis, 100 mL of distilled/deionized (DDI) water was added to a 0.5 g (oven dry equivalent) of P-source to give a 1:200 solid/solution ratio. High solid/water ratios reportedly relate better to runoff potential than lower solid/water ratios (Sharpley and Moyer, 2000; Brandt et al., 2004; O'Connor and Elliott, 2006). Samples were shaken on a reciprocating shaker at a rate of 100 strokes min–1 for 1 h at room temperature (24 ± 2°C). The suspensions were then centrifuged at a relative centrifugal force of 8000 x g for 15 min, and the supernatant filtered through 0.45-µm filter paper. Phosphorus concentrations in the extracts were determined by the Murphy and Riley (1962) colorimetric method. Five grams (oven-dry equivalent) of the P-sources were analyzed for total P, Fe, and Al by ICP-AES (PerkinElmer Plasma 3200, PerkinElmer, Wellesley, MA) following acid-peroxide digestion according to the USEPA Method 3050A (USEPA, 1986). The percentage of total P that is water-extractable (PWEP) was then calculated. Oxalate (200 mM) extractable P, Fe, and Al were determined by ICP–AES after extraction at a 1:60 solid/solution ratio, following the procedures of Schoumans (2000). We also calculated the environmentally effective P load for the various P sources applied. The environmentally effective P load term used in this paper refers to the product of PWEP and TP of the P source applied, and conceptually represents the fraction of total applied P available for transport. The P saturation index (PSI = [Pox]/[Alox + Feox]) was calculated from the molar concentrations of oxalate-extractable P (Pox), Al (Alox), and Fe (Feox) in the biosolids (Elliott et al., 2002). The PSI of P-source is a measure of P retention/release potential from a particular P-source. Since only Feox and Alox, together with Pox, are considered in the calculation of PSI, the index is not applicable for Ca-dominated amendments, such as TSP.

Rainfall Simulation Experiment
Rainfall simulation was performed as prescribed in the U.S. National Phosphorus Research Project indoor runoff box protocol (National Phosphorus Research Project, 2001) to assess P losses in runoff and leachate. The prescribed 100 cm long, 20 cm wide, and 7.5 cm deep runoff boxes were modified to quantify P leaching in addition to P runoff by adding a second (empty) water proof box under the first in a double-deck design (Agyin-Birikorang et al., 2007a). The top boxes were each packed with 5 cm (14 kg) of soil to a bulk density of 1400 kg m–3 and the entire double-deck box assemblage sloped to 3% during the rainfall events.

The P-sources were applied uniformly to the soil at an equivalent total P application rate of 224 kg P ha–1, corresponding to a typical biosolids application rate based on crop N requirements. Amendments were either surface-applied or incorporated into the entire 5-cm depth of soil. Amended soils were pre-wetted to near saturation to control for antecedent moisture and to promote runoff and leaching in the subsequent rainfall simulation. Rainfall simulations were conducted three times, at 1-d intervals between rainfall events, with rainfall delivered at 7.1 cm h–1 from a height of 3 m above the boxes as prescribed in the U.S. National Phosphorus Research Project indoor runoff box protocol. For each rainfall event and box, 30 min of runoff was generated by the simulated rainfall, and the volumes recorded. Simultaneously, leachate generated during the entire rainfall event was collected, and the volumes recorded. Runoff subsamples were immediately filtered (0.45 µm) for soluble reactive P (SRP) and total dissolved P (TDP) analyses. Representative well-mixed samples of the unfiltered runoff and leachate (~250 mL each) were also obtained from each replicate for additional analysis.

Leachate and runoff (filtered and unfiltered samples) pH and EC values were determined. Soluble reactive P was determined on the filtered runoff and the leachate samples colorimetrically (Murphy and Riley, 1962). Total dissolved P was measured on the filtered runoff and the leachate samples after digesting 10 mL of the samples with 0.5 mL 5.5 M H2SO4 and 0.15 g of potassium persulfate in an autoclave for 1 h (Pote and Daniel, 2000a, 2000b). Total P in the unfiltered runoff samples was determined by digesting 5 mL of the samples with 1 mL of 5.5 M H2SO4 and 0.3 g of potassium persulfate on a digestion block and then diluting to 10 mL with DDI water (Pote and Daniel, 2000b). All digested samples were analyzed for P colorimetrically (Murphy and Riley, 1962). The iron-oxide impregnated paper strip method (Myers and Pierzynski, 2000) was used to estimate bioavailable P in unfiltered runoff. Particulate phosphorus (PP) was calculated by subtracting TDP from the TP of each runoff sample. Flow-weighted P concentrations (SRP, TDP, or TP) were determined for the runoff and the leachate samples by summing the product of the P concentrations and volumes for the three runs to yield cumulative P mass, and dividing the P mass by the total volume of the runs. The masses of runoff and of leachate P losses (mg) were calculated as the product of flow-weighted concentrations (mg L–1) and the runoff and leachate volumes (L), respectively.

Quality Control
All sample collection/handling/chemical analysis was conducted according to a standard QA/QC protocol (Kennedy et al., 1994). For each set of samples, a standard curve was constructed (r2 > 0.998). Method reagent blanks, as well as certified standards from a source other than normal calibration standards, were included in the extraction process. Percentage recovery ranged from 97 to 103% of the expected values. A 5% matrix spike of the set was used to determine the accuracy of the data obtained, with recoveries ranging from 96 to 103% of the expected values. Another 5% of the set was used to determine the precision of the measurements (triplicates). Analyses that did not satisfy this QA/QC protocol were re-extracted and rerun.

Statistical Analyses
We tested the data for analysis of variance (ANOVA) assumptions (constant variance and normality using residual plots and the normal probability plots, respectively), with the SAS software (SAS Institute, 2002). Test results prompted us to transform all runoff and leachate data logarithmically to conform to ANOVA assumptions of normality and equal error variances. Treatment effects were evaluated using the general linear model (PROC GLM) of the SAS software version 9 (SAS Institute, 2002), along with Tukey's mean separation. Treatment differences were tested at a significance ({alpha}) level of 0.05.


   Results and Discussion
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Materials and Methods
 Results and Discussion
Summary and Conclusions
 REFERENCES
 
Phosphorus-Source Properties
Selected properties of the five biosolids and the TSP utilized for the study are presented in Table 2. Except for the Disney biosolids, all biosolids had total N levels ≥60 g kg–1, typical of biosolids produced nationally. The Disney biosolids had the lowest total N content (22 g kg–1), but such a low total N value is not uncommon for composted biosolids materials (USEPA, 1995; O'Connor and Elliott, 2006). Total P values of all the five biosolids were within the range of values (~8–40 g kg–1; USEPA, 1995; O'Connor and Elliott, 2006) reported for biosolids produced nationally. Total concentrations of major elements (Al, Fe, and Ca) were also representative of biosolids produced nationally, and reflected individual wastewater and sludge treatment processes. Thus, Fe or Al concentrations were generally ≤10 g kg–1, unless chemicals were added to the waste stream for P removal (e.g., Disney and Milorganite biosolids).

Biosolids varied widely in the amount of labile P estimated as P extracted with water (WEP). The two biosolids stabilized with Fe salts (Disney and Milorganite) had the lowest WEP values. Water-extractable P was shown in previous work (Brandt et al., 2004; Elliott et al., 2005, 2006; O'Connor and Elliott, 2006) to reflect reasonably the relative bioavailable and leachable P for most biosolids. The critical role of Al and Fe in determining P solubility and release from biosolids-amended soils has been documented by several researchers (Maguire et al., 2001; Brandt and Elliott, 2003; Shober and Sims, 2003; Elliott et al., 2005). Jokinen (1990) assessed the influence of treatment process on extractable soil P in biosolids-amended soils and concluded that although Al- and Fe-stabilized sludges tended to reduce P solubility, Ca-sludge increased soluble P values in soils. Soon and Bates (1982) reported that the extractability of P from soils amended with chemically treated, anaerobically digested sludges followed the order: Ca-sludge > > Fe-sludge > Al-sludges. The WEP content of manures and biosolids treated with Al and Fe salts or by-products is reported to be lower than their untreated counterparts (Codling et al., 2000; Elliott et al., 2002). Penn and Sims (2002) found that biosolids containing high amounts of Fe consistently had the lowest WEP and exhibited the lowest runoff dissolved P when added to soils.

We also calculated the PSI to estimate the degree to which biosolids P is potentially bound with Fe and Al. Elliott et al. (2002) suggested that PSI values <1 indicate excess Fe and Al for binding of P (little available P), whereas values >1 suggest available P beyond that associated with Fe and Al precipitates. Obviously, P-binding soil components will affect P availability once biosolids are land applied (Elliott et al., 2002), but the PSI value of biosolids can be useful as an index of biosolids-P lability in soils with low P sorption capacities. The PSI parameter qualitatively identifies the BPR and BPR-like biosolids materials (Orange County, Gainesville, and Lakeland biosolids) as highly labile P sources, and the non-BPR biosolids (Disney and Milorganite biosolids) materials as poorly labile P sources in poorly P-sorbing soils.

Our analysis confirmed the total P levels of the TSP fertilizer reported by the manufacturer. Much of the P in the TSP (~80%) was water-extractable as expected for a commercial fertilizer, and Ca was the dominant cation (140 g kg–1). Impurities in the rock phosphate treated with H3PO4 to produce the fertilizer contributed significant amounts of Fe and Al (Table 2).

Amendment Effects on Runoff and Leachate Phosphorus
A significant effect of the P sources on runoff and leachate P losses was observed for both methods of P-source application. For the surface-applied treatments, the TSP amended soil had the greatest amount (runoff + leachate) of TP and TDP loss (Fig. 1A , 1B), consistent with the high P solubility of the fertilizer. Although the biosolids were applied at the same total P rate (224 kg ha–1), the amounts of runoff + leachate TP and TDP loss were significantly (p < 0.05) different among the biosolids-treated soils. The greatest amount of TP loss among the biosolids occurred in the BPR and BPR-like biosolids (Lakeland biosolids treatment followed by Gainesville biosolids treatment and Orange County biosolids treatment in that order, Fig. 1A). The TP and TDP losses from the non-BPR biosolids (Disney and Milorganite biosolids) amended soils were similar, and significantly (p < 0.05) smaller than the amounts of P losses from the other biosolids (Fig. 1A, 1B), but were greater (p < 0.05) than that of the control treatment. The total amounts (runoff + leachate) of SRP loss from the surface-applied P source treatments followed similar trends as the TP and TDP losses, with the TSP treatment having the greatest SRP loss (Fig. 1C). However, unlike the TP and TDP data, the total mass of SRP loss from the Disney and Milorganite biosolids-treated soil samples were similar to that of the control treatment (Fig. 1C).


Figure 1
View larger version (23K):
[in this window]
[in a new window]

  		Fig. 1. Effects of surface-applied amendments on total amounts (runoff and leachate for three rainfall events) of (A) total P, (B) total dissolved P, and (C) soluble reactive P losses in runoff and leachate. Bars represent average values of three replicates. Treatments having the same letter are not different by the Tukey multiple comparison at significance level ({alpha}) of 0.05. CTRL = control samples (no P application); DSNY = Disney biosolids-treated soil; GRU = Gainesville biosolids-treated soil; LKLD = Lakeland biosolids-treated soil; mLGT = Milorganite biosolids-treated soil; OCUD = Orange County biosolids-treated soil; TSP = triple superphosphate-treated soil.

 
Several studies have shown that the potential loss risk of P in BPR and BPR-like biosolids is greater than their non-BPR counterparts (Elliott et al., 2002, 2005, 2006; O'Connor and Elliott, 2006). As a category, BPR biosolids have the greatest WEP values among biosolids types (Brandt et al., 2004). Besides metal salt additions, other processes used in wastewater treatment and biosolids generation (e.g., heat-drying) influence biosolids P solubility and runoff potential (Penn and Sims, 2002; Brandt et al., 2004). O'Connor and Sarkar (1999) suggested that the relatively low bioavailability of heat-treated (pelletized) biosolids may be partly attributed to slow physical breakdown. Under field conditions, pellets were still distinguishable 5 mo after soil application. Shaking pellets for extended times, or grinding the pellets, increased potassium chloride-extractable phosphorus. Thus, the data provide strong evidence that the solubility of applied P sources exerts a major influence on the potential for offsite migration of P at land application sites. In an evaluation study of the P indices of 47 states in the United States, Sharpley et al. (2003) observed that only nine states, including Florida, have incorporated an estimate of solubility of P applied to the site. The current Florida P index use a single coefficient (0.015) for all types of biosolids (Graetz et al., 2004), although proposed revisions are being considered that will differentiate biosolids based on the P solubility of the biosolids (Willie G. Harris, personal communication, 2007). Phosphorus source coefficients (PSC) have been developed as quantitative indicators of the relative solubility of the P in land-applied materials. Several other studies (Leytem et al., 2004; Elliott et al., 2006; O'Connor and Elliott, 2006) confirm the need to incorporate weighting coefficients (PSC values) into P site indices to reflect the portion of total applied P susceptible to off-site transport (total P added x PSC).

As with the surface application treatments, significant P-source effects were observed when the materials were incorporated into the soil. As expected, the greatest amounts of P loss (TP, TDP, and SRP), occurred in the TSP treatment (Fig. 2 ). Phosphorus losses among the biosolids were significantly different. The Lakeland biosolids treatment had the greatest amounts of P loss among the biosolids. Unlike the results of the surface application study, the amounts of runoff + leachate TP loss from the incorporated Gainesville and Orange County biosolids-treated soil samples were similar, but were greater than those of Disney and Milorganite treatments. The total masses of dissolved P (TDP and SRP) losses in the incorporated treatments were similar among the Disney, Milorganite, and the control treatments (Fig. 2), as in the surface application treatment.


Figure 2
View larger version (24K):
[in this window]
[in a new window]

  		Fig. 2. Effects of soil incorporation of P sources on total amounts (runoff and leachate for three rainfall events) of (A) total P, (B) total dissolved P, and (C) soluble reactive P losses in runoff and leachate. Bars represent average values of three replicates. Treatments having the same letter are not different by the Tukey multiple comparison at significance level ({alpha}) of 0.05.

 
Effects of Phosphorus-Source Application Methods on Off-Site Phosphorus Loss
The surface-applied treatment mimicked spreading of the P sources in the field without incorporating, whereas the P-source incorporation method mimicked field application followed by thorough mixing in the root zone of soils. Analysis of variance showed that there was no significant difference (p > 0.05) between the respective total masses (runoff + leachate) of SRP and TDP loss from similar P-source treatments either surface-applied or incorporated into the Immokalee soil (Fig. 1B vs. 2B, and 1C vs. 2C). The absolute values of the total (runoff + leachate) amounts of TP losses from some of the surface-applied BPR and BPR-like biosolids (Lakeland, Gainesville, and Orange County) materials were greater than those of similar treatments when the P sources were incorporated into the soil, but the differences were not statistically significant (Fig. 1A vs. 2A). The slight increase in the total masses of TP losses from the surface application of the three BPR and BPR-like biosolids materials resulted from greater amounts of particulate P (PP) losses occurring in the surface application treatments than soil incorporation method. The slurry nature of the three BPR and BPR-like biosolids materials (Lakeland biosolids had ~4% solids, Gainesville biosolids had 5% solids, and Orange County biosolids had 16%, Table 2) promoted sealing of the soil surface after application, and surface erosion. Thus, the PP was likely derived primarily from the P sources rather than from the soil. The greater runoff PP observed for the three BPR and BPR-like biosolids materials when surface-applied (than incorporated) prompted estimation of BAP levels in the runoff water using the iron-oxide impregnated paper strip method (Myers and Pierzynski, 2000). Sharpley (1993) reported that the transport of BAP in agricultural runoff can stimulate freshwater eutrophication. The total masses of runoff BAP from the surface-applied P sources (sum of the three rainfall events), were similar to masses of runoff TDP of corresponding P sources (Fig. 3 ), suggesting that the solubility of the PP in runoff was low.


Figure 3
View larger version (25K):
[in this window]
[in a new window]

  		Fig. 3. Runoff bioavailable P (BAP) and total dissolved P (TDP) losses arising from the surface application of the P sources for the three rainfall events. Bars represent average values of three replicates. Treatments having the same letter are not different by the Tukey multiple comparison at significance level ({alpha}) of 0.05. CTRL = control samples (no P application); DSNY = Disney biosolids-treated soil; GRU = Gainesville biosolids-treated soil; LKLD = Lakeland biosolids-treated soil; mLGT = Milorganite biosolids-treated soil; OCUD = Orange County biosolids-treated soil; TSP = triple superphosphate-treated soil.

 
For the surface-applied treatments, total masses of the various forms of P (SRP, TDP, and TP) lost among the P sources were dominated by losses via surface runoff (Fig. 4 ), but P losses via leaching were significant. For example, the amount of leachate TP loss from the surface-applied P sources was ~35% of the total (runoff + leachate) amount of TP loss (Fig. 4A). In contrast, P loss through leaching was the dominant P loss mechanism when the P sources were incorporated into the soil (Fig. 4). The leachate TP losses from incorporated P sources accounted for ~75% of the total (runoff + leachate) mass of TP loss (Fig. 4A). The total amounts of SRP and TDP losses from the P sources followed similar trends (Fig. 4B, 4C) as mass of TP loss (Fig. 4A). Total amounts of SRP and TDP lost from the surface-applied P sources were significantly greater in runoff than in leachate, although leachate SRP and TDP losses were ~30% and 40%, respectively, of the total (runoff + leachate) amount of SRP and TDP losses (Fig. 4B, 4C). Contrary to the surface-applied treatments, SRP and TDP losses were dominated by losses through leaching when the P sources were incorporated. Runoff SRP and TDP losses accounted for <25% of the corresponding total (runoff + leachate) amounts of SRP and TDP losses from the P sources. Thus, contrary to results obtained with strongly P-retaining soils (Peterson et al., 1994; Spargo et al., 2006; Elliott et al., 2005), P leaching is a major P loss mechanism in the poorly P-retaining Immokalee soil, and cannot be ignored. Several studies have shown that off-site P losses via leaching in Florida Spodosols (and other southeastern coastal plain soils) is a matter of concern (Elliott et al., 2002; O'Connor et al., 2005; Alleoni et al., 2008). Depending on the topography and surface condition of the field (presence or absence of surface vegetation), P leaching can be more important than P losses via runoff (O'Connor et al., 2005).


Figure 4
View larger version (20K):
[in this window]
[in a new window]

  		Fig. 4. Effects of amendment application method on P losses in runoff and leachate in the form of (A) total P, (B) total dissolved P, and (C) soluble reactive P. Bars represent average values of 18 replicates. Treatments having the same letter are not different by the Tukey multiple comparison at significance level ({alpha}) of 0.05.

 
Although the depth of boxes utilized for the rainfall simulation experiment (7.5 cm) is much less than depths of typical soil profile of naturally occurring Spodosols, the results obtained from the study can still guide management of P transport in poorly P-sorbing soils. Several studies have shown that Florida Spodosols have surface horizons that sorb P poorly, overlying an E-horizon with practically no P sorption capacity, but contain an Al-rich spodic horizon (Graetz and Nair, 1995, Nair et al., 1995; Harris et al., 1996). One study (Agyin-Birikorang et al., 2007b) reported an oxalate-extractable Al + Fe concentration of ~40 mmol kg–1 in the spodic horizon of a typical Florida Spodosol. Aluminum and Fe oxides are major sorbent of P (Dayton and Basta, 2005; Elliott et al., 2005; Agyin-Birikorang et al., 2007a), thus the spodic horizon could retain P leached from the surface horizon before reaching the aquifer. However, the hydrology of most Florida Spodosols (high water tables in rainy seasons) results in leached P being intercepted by ground water before coming into contact with the spodic horizon (Graetz and Nair, 1995; Harris et al., 1996). The P-contaminated ground water then moves to surface water via lateral subsurface flow (Burgoa et al., 1991; Mansell et al., 1991; Harris et al., 1996). Thus, P lost in leachate from the boxes of the rainfall simulation experiment could be a good measure of the potential P problems in Florida Spodosols. Although the current Florida P index emphasizes management practices that minimize surface runoff to reduce field site P vulnerability, the data clearly suggest that controlling off-site P losses in Florida Spodosols require efforts to prevent or minimize P leaching, as well as runoff P losses.

Despite the greater runoff PP occurring in the surface application of the three BPR-like biosolids materials than soil incorporation, the sums (runoff + leachate) of the TP losses were similar whether the P sources were surface-applied or incorporated. Data from this study suggest that in poorly P-sorbing soils, the method of P-source application determines the mechanism of P loss (either through runoff or leachate), but minimally affects the total amount of P loss resulting from the applied P source. Therefore, the promotion of P-source incorporation to reduce off-site P movement may not be justifiable for poorly P-sorbing soils. Incorporation merely converts surface runoff losses to leaching losses.

Relationship between Dissolved Phosphorus and the Environmentally Effective Phosphorus Load
Figure 5 shows the relationship between TDP of cumulative runoff and leachate mass and (A) WEP and (B) the environmentally effective load of the P sources used for the study. The data presented in Fig. 5A and 5B are combined values for the two P application methods, since similar cumulative masses of P lost (runoff + leachate for the three rainfall events) were observed for individual P sources.


Figure 5
View larger version (14K):
[in this window]
[in a new window]

  		Fig. 5. (A) Concentration of total dissolved P as a function of source water-extractable P (WEP), and (B) mass of total dissolved P as a function of environmentally effective P load of the P sources (PWEP x total P applied).

 
A positive correlation between TDP loss and application rate (Kleinman and Sharpley, 2003) is logical for a given P-source type because the amount of soluble P available to runoff and percolating water depends on the application rate. Such a relationship should not necessarily hold for a broad range of P sources where the potential to release P into runoff and leaching water is highly variable (Elliott et al., 2006). There was no correlation between TDP and the total P application rate in our study, consistent with the observation of Elliott et al. (2006) and O'Connor and Elliott (2006). Our study consisted of six P sources applied at the same total P rate (224 kg ha–1), but the amounts of runoff + leachate TDP lost ranged from ~27 to ~700 mg (Fig. 1B and 2B). Others have also shown that biosolids applied at the same total P rate exhibit significantly different runoff P losses (Penn and Sims, 2002; Brandt and Elliott, 2003; Sims et al., 2003). Elliott et al. (2002) and Brandt et al. (2004) showed that runoff P concentrations were related to the WEP levels of biosolids and manures. For the wide range of materials utilized in their studies, WEP and TDP were linearly related (r2 > 0.8). The data (Fig. 5A) are described reasonably well by a linear equation [TDP (mg L–1) = 2.1 + 2.3 x WEP (g P kg–1), r2 = 0.9]. The progressive increase of masses of TDP loss with P-source WEP has been widely documented, and serves as the basis for the use of WEP as a quantitative indicator of the potential for manures and biosolids to release dissolved P to runoff (Kleinman et al., 2002). Some states (e.g., Pennsylvania and Arkansas) use P-source WEP concentrations as indicator of P loss potential in site assessment indices (Weld et al., 2000). In Pennsylvania, the WEP values have been used to develop P source coefficients (PSCs) as quantitative indicators of the relative availability of the P in land-applied materials to be transported in runoff (Elliott et al., 2006). These PSCs are fractional values referenced to mineral fertilizer (i.e., fertilizer PSC = 1.0). In the Pennsylvania P index, PSCs are multiplied by the application rate of the organic sources of P to obtain the fraction of total applied P available for transport (Elliott et al., 2006). As P-based nutrient management is implemented for land application, it is imperative that guidance and regulatory protocols reflect the substantially different solubilities and mobilities of P in P-sources. Total P levels are not useful for such purposes. Water-extractable P is easily obtained and seems a superior measure of the environmentally relevant portion of P in biosolids.

Cumulative runoff + leachate TDP masses of both application methods were linearly and positively correlated with PWEP values of the P sources (data not presented). In a similar study, Alleoni et al. (2008) showed that runoff dissolved P was exponentially related to PWEP. Brandt et al. (2004) also reported that PWEP of manures and biosolids decreases with increasing total Al and Fe content of the materials, and the trend has been modeled as a decreasing power function. Chinault (2007) showed that biosolids with PWEP values ≥14% have a larger potential negative environmental impact than biosolids with PWEP values <14%. O'Connor and Elliott (2006) and Brandt et al. (2004) observed that PWEP of a particular P-source is strongly correlated with PSI of the source. Hence, PSI can also be used to gauge the environmental impact of biosolids land application to sandy soils with minimal P-sorbing capacity. When the PWEP was multiplied by the total P applied (PWEP x Total P load; referred to in this article as environmentally effective P load), a stronger linear relationship with the total (runoff + leachate) TDP [TDP (g) = y = 0.0192 +0.0048x (where x is the environmentally effective P load in kg ha–1), r2 = 0.94] was observed (Fig. 5B).The environmentally effective P load is the estimate of the portion of the applied P potentially susceptible to off-site movement to cause environmental problems. The data clearly suggest that environmentally effective P load could be used to predict environmental hazard likely to result from applied P sources in poorly P-sorbing soils. Therefore, measurements such as WEP, PWEP, and the environmentally effective P load should be considered in regulating biosolids land application. O'Connor and Elliott (2006) concluded that "biosolids land application should not be regulated by assuming all biosolids have equal amounts of labile P. Assuming all biosolids to have equal amounts of labile P, and requiring P-based application rates without considering a residual's individual environmental hazard would unfairly burden municipalities facing disposal problems."


   Summary and Conclusions
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Materials and Methods
 Results and Discussion
 Summary and Conclusions
REFERENCES
 
This study was conducted using surface horizon material of a typical Florida Spodosol (Immokalee fine sand, sandy siliceous hyperthermic Arenic Alaquods) with low P sorption capacity. We utilized simulated rainfall to investigate effects of P-source characteristics and application methods on the forms and amounts of P losses from six P sources, including five biosolids produced and/or marketed in Florida. Total amounts of P losses from individual P sources were similar, whether the amendment was surface-applied or incorporated into the soil. Masses of SRP, TDP, and TP losses ranged from 21 to 550, 27 to 600, and 29 to 640 mg, respectively, among the P sources. A strong linear (r2 = 0.94) and significant (p < 0.001) positive correlation was observed between dissolved P loss and the environmentally effective P load of the P sources, which is calculated as the product of PWEP of a P source and the TP load applied to the soil. We conclude that in the Immokalee fine sand (and likely other soils with low P sorption capacity) the method of P-source application determines the mechanism of P loss (either through runoff or leachate), but not the total amount of P loss resulting from the applied P-source. Rather, the solubility of the amendment determines the P loss potential of the P-source, and P solubility indices (such as PWEP) should be included in estimates of P loss potential of biosolids and other residuals. Controlling off-site P losses in Florida Spodosols requires best management efforts other than P-source application methods. Soil amendments that have the potential to increase sorption capacity of poorly P-sorbing soils should be applied to such soils to minimize P leaching as well as runoff P losses.


   ACKNOWLEDGMENTS
 
This study was funded by the Florida Water Environmental Association (FWEA).


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 Materials and Methods
 Results and Discussion
 Summary 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.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Materials and Methods
 Results and Discussion
 Summary and Conclusions
 REFERENCES
 




This article has been cited by other articles:


Home page
 Agron. J.Home page
Y. C. Newman, S. Agyin-Birikorang, M. B. Adjei, J. M. Scholberg, M. L. Silveira, J. M. B. Vendramini, J. E. Rechcigl, L. E. Sollenberger, and M. Chrysostome
Enhancing Phosphorus Phytoremedation Potential of Two Warm-Season Perennial Grasses with Nitrogen Fertilization
Agron. J., November 1, 2009; 101(6): 1345 - 1351.
[Abstract] [Full Text] [PDF]

Home page
 Agron. J.Home page
Y. C. Newman, S. Agyin-Birikorang, M. B. Adjei, J. M. Scholberg, M. L. Silveira, J. M. B. Vendramini, J. E. Rechcigl, and L. E. Sollenberger
Nitrogen Fertilization Effect on Phosphorus Remediation Potential of Three Perennial Warm-Season Forages
Agron. J., August 31, 2009; 101(5): 1243 - 1248.
[Abstract] [Full Text] [PDF]

Home page
 Agron. J.Home page
M. Miller and G. A. O'Connor
The Longer-Term Phytoavailability of Biosolids-Phosphorus
Agron. J., July 7, 2009; 101(4): 889 - 896.
[Abstract] [Full Text] [PDF]

Home page
 Soil Sci.Home page
S. Sato, K. T. Morgan, M. Ozores-Hampton, and E. H. Simonne
Spatial and Temporal Distributions in Sandy Soils with Seepage Irrigation: II. Phosphorus and Potassium
Soil Sci. Soc. Am. J., May 1, 2009; 73(3): 1053 - 1060.
[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
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Agyin-Birikorang, S.
Right arrow Articles by Brinton, S. R.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Agyin-Birikorang, S.
Right arrow Articles by Brinton, S. R.
Agricola
Right arrow Articles by Agyin-Birikorang, S.
Right arrow Articles by Brinton, S. R.
Related Collections
Right arrow Phosphorus
Right arrow Residue management
Right arrow Soil Chemistry
Right arrow Other Soil Management
Right arrow Municipal Waste

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
Soil Science Society of America Journal Journal of Plant Registrations The Plant Genome