JEQ Grow Your Career With ASA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (15)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Börling, K.
Right arrow Articles by Barberis, E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Börling, K.
Right arrow Articles by Barberis, E.
Agricola
Right arrow Articles by Börling, K.
Right arrow Articles by Barberis, E.
Related Collections
Right arrow Batch Studies
Right arrow Sorption/Exchange
Right arrow Ecological Risk Assessment
Right arrow Phosphorus
Right arrow Soil Analysis
Published in J. Environ. Qual. 33:99-106 (2004).
© ASA, CSSA, SSSA
677 S. Segoe Rd., Madison, WI 53711 USA

TECHNICAL REPORTS

Ecological Risk Assessment

Soil Variables for Predicting Potential Phosphorus Release in Swedish Noncalcareous Soils

Katarina Börling*,a, Erasmus Otabbonga and Elisabetta Barberisb

a Department of Soil Sciences, Swedish University of Agricultural Sciences, P.O. Box 7014, SE-750 07 Uppsala, Sweden
b Dipartimento di Valorizzazione e Protezione delle Risorse Agroforestali, University of Turin, Via Leonardo da Vinci 44, I-10095 Grugliasco (TO), Italy

* Corresponding author (Katarina.Borling{at}mv.slu.se).

Received for publication February 12, 2003.

   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
The accumulation of P in agricultural soils due to fertilization has increased the risk of P losses from agricultural fields to surface waters. In risk assessment systems for P losses, both P release from soil to solution and transport mechanisms need to be considered. In this study, the overall objective was to identify soil variables for prediction of potential P release from soil to solution. Soils from nine sites of the Swedish long-term fertility experiment were used, each with four soil P levels. Phosphorus extractable with CaCl2 was used as an estimate of potential P release from soil to solution. Ammonium lactate–extractable phosphorus (P-AL) or NaHCO3–extractable phosphorus (Olsen P) could not be used alone for prediction of potential P release since soils with high phosphorus sorption capacity (PSC) released less P than soils with low PSC at the same soil test phosphorus (STP) level. Degree of phosphorus saturation (DPS) was calculated as Olsen P or P-AL as a percentage of PSC derived from P sorption isotherms or from Fe and Al extractable in ammonium oxalate. The CaCl2–extractable total phosphorus (CaCl2–TP) was exponentially related to these DPS values (r2 ≥ 0.79). The CaCl2–TP was also linearly related to ratios between Olsen P or P-AL and a single-point phosphorus sorption index (PSI; r2 ≥ 0.86). These ratios, which are easily determined and gave good correlations with CaCl2–TP, seemed to be the most useful estimates of potential P release for risk assessment systems.

Abbreviations: Alox, Feox, and Pox, ammonium oxalate–extractable aluminum, iron, and phosphorus, respectively • CaCl2–RP, CaCl2–extractable reactive phosphorus • CaCl2–TP, CaCl2–extractable total phosphorus • CaCl2–UP, CaCl2–extractable unreactive phosphorus • DPS, degree of phosphorus saturation • EPC, equilibrium phosphorus concentration • Olsen P, NaHCO3–extractable phosphorus • P-AL, ammonium lactate–extractable phosphorus • PSC, phosphorus sorption capacity • PSCmax, maximum phosphorus sorption capacity • PSCox, phosphorus sorption capacity calculated using the equation of Börling et al. (2001) • PSI, phosphorus sorption index • STP, soil test phosphorus


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
THE AVERAGE PHOSPHORUS ACCUMULATION in Swedish arable topsoils since the 1950s has been estimated at 600 to 700 kg P ha–1 (Andersson et al., 1998). This is a result of fertilization both with mineral fertilizers and manure in excess of plant needs. This accumulation increases the risk of P losses from agricultural fields to surface waters, causing eutrophication (Sims et al., 1998). The main transport pathways for P losses from agricultural fields are erosion, runoff, and subsurface leaching, and P can be transported both in particulate and soluble forms (Sims et al., 1998). It is well known that P from agricultural soils contributes significantly to the P load of surface waters in Sweden (Larsson, 1997; Ulén, 1997), and subsurface leaching has been reported to contribute significantly to P losses (Ulén, 1997). It is therefore necessary to identify soils vulnerable to P leaching losses. In risk assessment systems for P losses, both transport mechanisms and P source need to be considered (Gburek et al., 2000). When considering P source, soil P status or potential P release from soil to solution must be considered, since release of P to the soil solution is crucial for losses of dissolved P (Heckrath et al., 1995).

Agronomic soil tests for P, such as Olsen P or Mehlich-3 P, are often used as an estimate of P source in risk assessment as large amounts of data are available for these methods (Sims et al., 2000). Significant correlations between STP values and P release to drainage water have been reported for individual soils (e.g., Heckrath et al., 1995; Hesketh and Brookes, 2000). However, critical STP values above which P release increases significantly can differ between soils (Hesketh and Brookes, 2000).

The soil PSC, which also influences P release, can be used in addition to STP to predict P release from soil to solution (van der Zee et al., 1990; Sharpley, 1995). The ratio of sorbed P to PSC of the soil can be estimated. If the parameters are expressed in the same units, it is possible to estimate the DPS in the soil:

[1]

Degree of P saturation estimates how close the soil is to saturation (Sharpley, 1995) and provides possibilities to compare soils with varying P sorption capacities. It can thus be useful as an estimate of P source in risk assessment systems (van der Zee et al., 1990; De Smet et al., 1996).

Sorbed P is usually estimated using P extracted with acid ammonium oxalate solution (van der Zee and van Riemsdijk, 1988), anion exchange resin P (Sibbesen, 1978), iron-impregnated strips (van der Zee et al., 1987), or different STP methods (Sharpley, 1995; Kuo et al., 1988; Kleinman et al., 2000). The PSC can be determined from P sorption isotherms and calculations based on the Langmuir or Freundlich equations. However, some indirect methods, which are less laborious, can also be used to estimate PSC. One is based on extraction of iron (Fe) and aluminium (Al) in acid ammonium oxalate solution (Schwertmann, 1964). Another is based on a single point addition of P (Bache and Williams, 1971) from which a PSI is calculated. The PSI has been shown to be a good estimate of PSC in Swedish soils (Börling et al., 2001). Different methods of estimating sorbed P and PSC can be combined to determine DPS. In the Netherlands and several other countries, P extracted with ammonium oxalate has been used to estimate sorbed P, while Fe and Al in the same extract are used to estimate PSC. Sharpley (1995) used Melich-3 P or iron-impregnated-strip P to estimate sorbed P and sorption isotherms to estimate PSC.

The method to be used to estimate sorbed P and PSC for determination of DPS in a region or a country needs to be tested on representative soil types. Furthermore, it needs to be well correlated with measures of P concentration in the soil solution or in runoff or drainage waters (an intensity factor). Some authors measure P concentration in tile drains or runoff (e.g., Heckrath et al., 1995; Sharpley, 1995) or in lysimeter experiments (e.g., Hesketh and Brookes, 2000) while others estimate P concentration by extracting P in water to simulate runoff (e.g., McDowell and Sharpley, 2001a, 2001b) or in CaCl2 to simulate P in soil solution (e.g., Hesketh and Brookes, 2000; McDowell and Sharpley, 2001a).

The objective of this study was to identify soil variables suitable for prediction of potential P release from soil to solution to be used as a tool in a risk assessment for P leaching losses. Two soil tests for P were studied, Olsen P and P-AL, which is the most common agronomic method used in Sweden. Since few studies have been conducted on P saturation in Swedish soils, DPS or the ratio of sorbed P and an index of PSC were calculated and evaluated by combining several different methods of estimating sorbed P and PSC. In this study, P extractable in CaCl2 was used as an estimate of P release from soil to solution.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Site Description and Soil Sampling
Soil samples were collected in autumn 1997 at nine sites of the Swedish long-term field experiment. Five sites are located in southern Sweden (Fjärdingslöv, S. Ugglarp, Örja, Ekebo, and Orup) and another four are located in central Sweden (Klostergården, Bjertorp, Högåsa, and Kungsängen). The former experiments were started in 1957 while the latter were started between 1963 and 1966. Treatments consist of four phosphorus–potassium (PK) fertilization levels (Table 1). Phosphorus fertilizer is applied as superphosphate and K fertilizer as muriate of potash. Each PK level is in combination with four nitrogen levels ranging from 0 to 150 kg N ha–1 yr–1. Nitrogen is applied as calcium ammonium nitrate. More details on the experimental design and yield response to fertilization are reported by Carlgren and Mattsson (2001).


View this table:
[in this window]
[in a new window]
  		Table 1. Fertilizer treatments in the field experiment.

 
In all PK levels, at least seven subsamples were taken at a 0- to 20-cm depth and mixed to one composite sample. Nitrogen applications in the soils sampled are 82 to 125 kg N ha–1 yr–1. The samples were air-dried and sieved through a 2-mm sieve before analysis.

Soil Analysis
Texture was determined by the pipette sedimentation procedure as described by Ljung (1987). Soil pH was measured in water at a soil to water ratio of 1:5. Organic carbon (C) was determined by dry combustion in an E-12 CNS 2000 carbon analyzer (LECO, St. Joseph, MI). Olsen P was determined according to Olsen and Sommers (1982). The P-AL buffered solution (pH = 3.75) was determined as described by Egnér et al. (1960). The P-AL was also determined in soils sampled before the start of field experiments (initial samples) (Carlgren and Mattsson, 2001).

The ammonium oxalate–extractable iron, aluminium, and phosphorus (Feox, Alox, and Pox, respectively; pH = 3) were determined according to Schwertmann (1964). Iron and aluminium in the oxalate extract were analyzed by atomic absorption spectroscopy. Phosphorus was measured colorimetrically according to Wolf and Baker (1990). Phosphorus sorption isotherms were derived from sorption data of at least nine KH2PO4 additions per soil. Two grams of air-dried soil (≤2 mm) were equilibrated with 20 mL 0.01 M CaCl2 containing different amounts of P. Two drops of toluene were added to inhibit microbial activity. The tubes were shaken for 20 h at 21°C. Following centrifugation at 3000 rpm for 10 min, the equilibrium P concentration in the solution (C) was analyzed colorimetrically according to Murphy and Riley (1962). The amount of P sorbed by the soil (X), in mmol kg–1 soil, was calculated by subtracting the amount of P in the equilibrium solution from the amount added. Phosphorus sorption at a single-point addition of 19.4 mmol P kg–1 soil was also determined following the same procedure (Börling et al., 2001).

Soil samples were equilibrated with 0.01 M CaCl2 (1:3 w/v) at 23°C for 1 h (Schofield, 1955). Following centrifugation at 1500 rpm for 10 min, extracts were filtered through Millipore (Billerica, MA) filter paper (<0.45 µm) and P was subsequently fractionated. The CaCl2–RP in the filtered solution was determined according to the green malachite method (Ohno and Zibilske, 1991). The CaCl2–TP was determined after digestion in sulfuric–perchloric acid (Martin et al., 1999). In brief, a 4-mL portion of the filtrate was evaporated and dried at 105°C and 1 mL H2SO4 (18 M) and 0.5 mL HClO4 (12 M) were added to the dried matter. The samples were then digested at 200°C, cooled, and diluted to 10 mL with deionized water. Phosphorus in the digestate was determined according to the malachite green method. The CaCl2–extractable unreactive phosphorus (CaCl2–UP) was calculated as the difference between CaCl2–TP and CaCl2–RP.

Sorption Isotherms and Saturation Index
The values for Pox, Feox, and Alox were used for calculating DPS according to Eq. [2]:

[2]
with all parameters in mmol kg–1 and {alpha} = 0.5 as suggested by van der Zee et al. (1990).

The maximum phosphorus sorption capacity (PSCmax) was calculated from P sorption isotherms using the Langmuir equation in soil sampled from Level A, as these soils were not influenced by P excess (Börling et al., 2001). Also, equilibrium phosphorus concentration (EPC) was calculated for soils in Level A, using the lowest part of the sorption isotherm at P concentrations of 0 to 0.10 mg L–1, where the isotherm is usually linear (Hartikainen, 1991). The EPC was defined as the P concentration in the solution at which no net sorption or desorption occurred and it was calculated as the P concentration in the equilibrium solution when P sorbed equaled zero (Hartikainen, 1991). Phosphorus sorption capacity was also calculated using the equation suggested for Swedish soils by Börling et al. (2001):

[3]

Four different estimates of DPS (DPS-2 to DPS-5) were calculated by combining Olsen P or P-AL (recalculated in mmol P kg–1 soil) with the above-mentioned methods of estimating PSC (Table 4).


View this table:
[in this window]
[in a new window]
  		Table 4. Relationship between degree of phosphorus saturation (DPS) and CaCl2–extractable total phosphorus (CaCl2–TP).{dagger}

 
A single-point PSI was calculated as the amount of P sorbed by the soil (X), after a single-point addition of 19.4 mmol P kg–1 soil, divided by the logarithm of C (in µM) in the equilibrium solution (X/log C; Börling et al., 2001). The ratio of Olsen P or P-AL (recalculated in mmol P kg–1 soil) to PSI was calculated.

Statistical Analyses
Unless specified, all analyses were performed in triplicate. Soils were sampled from one replication in the field, which limits the ability to do statistical analysis on the differences between treatments at each site (which is not the main purpose of this study). However, differences between treatments in all soils were calculated with the nine sites as replicates and were separated by t test at 95% probability. The statistical analyses were performed using SAS (SAS Institute, 1999).


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Some general properties of the soils from Level A are presented in Table 2. The soils are arranged in order of increasing PSCmax, which ranged from 6.0 to 12.2 mmol kg–1 soil. Clay content ranged from 7 to 59%, pH ranged from 5.8 to 7.5, and organic carbon ranged from 1.1 to 2.3% in the soils. Content of Feox and Alox ranged from 30 to 68 mmol kg–1 soil, except for Kungsängen, an acid sulfate soil, where Feox was 151 mmol kg–1 soil.


View this table:
[in this window]
[in a new window]
  		Table 2. Some general properties, maximum phosphorus sorption capacity (PSCmax), and equilibrium phosphorus concentration (EPC) for Level A.

 
The pH, organic carbon, and Feox and Alox were also determined in Levels B, C, and D for all soils and no significant differences were found between the P levels for these properties (Table 3). As expected, Olsen P and P-AL increased significantly at each level from A to D (Table 3). Compared with P-AL values in the initial samples, taken at the beginning of the field experiment, P-AL was significantly less in Level A (Table 3), decreasing by 19 to 55 mg P kg–1 soil (Fig. 1) . The differences between Level B and the initial samples were not significant (Table 3). However, P-AL increased slightly in Fjärdingslöv while it was equal in two soils and decreased in the others for Level B (Fig. 1). This implies that replacement fertilization of P was not sufficient to maintain the P level in six out of nine soils. This was probably due to the retention of P in soils, caused by adsorption and precipitation reactions, which renders P less available (Sample et al., 1980). In Levels C and D, the soils were enriched in P, with P-AL increasing by as much as 153 mg P kg–1 soil, compared with the initial samples (Table 3; Fig. 1). The increases were greatest in Fjärdingslöv and least in Bjertorp. Target values of P-AL for crop production vary considerably between countries; for example, 41 to 80 mg kg–1 is considered optimal in Sweden for most crops (Jordbruksverket, 2000), while in Belgium 120 to 180 mg kg–1 is considered optimal (De Smet et al., 1996).


View this table:
[in this window]
[in a new window]
  		Table 3. Means for treatments in all soils.{dagger}

 


View larger version (28K):
[in this window]
[in a new window]
  		Fig. 1. Changes in ammonium lactate–extractable phosphorus (P-AL) as influenced by repeated cropping and fertilization. Level A, no P added; Level B, replacement of P; Level C, replacement + 15 or 20 kg P; Level D, replacement + 30 kg P.

 
The EPC, at which no net sorption or desorption occurred, was determined using the lowest part of the sorption isotherm that was linear (0.73 ≤ r2 ≤ 0.99) for all soils in Level A. Values of EPC obtained were equal to or less than 0.025 mg L–1. These values are low compared with those obtained by Hartikainen (1991) for five cultivated Finnish soils that did not receive any P application for seven years (EPC ≥ 0.043 mg L–1). This indicates that in Level A, very small amounts of P were sorbed to the soil surfaces in agreement with the low Olsen P values and the decreases in P-AL compared with the initial samples. Hence, soils from Level A were considered depleted in P.

Total P released by CaCl2 increased with increasing STP for all soils, but the increases differed markedly between soils (Fig. 2a, 2b) . Several authors have used this kind of relationship to calculate a change point in STP, above which P release increases more per unit STP than below (e.g., Heckrath et al., 1995). However, as this long-term field experiment only consisted of four different P levels for each soil, it was not possible to accurately calculate the change point. According to McDowell et al. (2001) at least eight points were needed to determine the change point with sufficient accuracy. Nevertheless, a visual estimate indicates that the change point was probably not reached in the four soils with greatest PSCmax (Högåsa, Ekebo, Kungsängen, and Orup) while the others had probably exceeded the change point in Levels C or D (Fig. 2a, 2b). The logarithm of CaCl2–TP was linearly correlated with Olsen P (r2 ≥ 0.77) and P-AL (r2 ≥ 0.74) for each soil. Soils with low PSC released more P into CaCl2 solution than soils with high PSC at the same STP level (Fig. 3a, 3b) . The CaCl2–RP was related to Olsen P and P-AL in an analogous way to CaCl2–TP (i.e., soils having the lowest P sorption capacities had the largest increase in CaCl2–RP as soil P status increased). The results obtained in this work were similar to those obtained by Sharpley (1995). The CaCl2–RP fraction represents mainly inorganic orthophosphate, though it may also contain small amounts of P hydrolyzed from organic substances or colloids (Sinaj et al., 1998; McDowell and Sharpley, 2001b). In the soils studied, reactive P was by far the main P form released at high CaCl2–TP, while CaCl2–UP, which represents organic P and inorganic polyphosphates (Ron Vaz et al., 1993) or colloids (Sinaj et al., 1998), was the dominant form in soils with low P status. The CaCl2–UP ranged from 1.7 to 96% of CaCl2–TP (Fig. 4) . A similar decrease in relative content of unreactive P in response to increasing total P was found by McDowell and Sharpley (2001b) and Ron Vaz et al. (1993). In contrast to CaCl2–RP and CaCl2–TP, CaCl2–UP was more or less constant from Levels A to D and was not correlated to STP. However, unreactive P measured after extraction of dried soil can partly consist of P released from lysis of microbial cells since microbes can be killed by osmotic shock after rapid rewetting (Turner and Haygarth, 2001). Thus, microbial P in the soil could be a more important factor for CaCl2–UP than STP. Hence, there are good relationships between STP and CaCl2–TP and CaCl2–RP, but these relationships are dependent on soil P sorption properties and are therefore site specific.



View larger version (31K):
[in this window]
[in a new window]
  		Fig. 2. The CaCl2–extractable total phosphorus (CaCl2–TP) in relation to soil test phosphorus (STP) as determined by (a) NaHCO3–extractable phosphorus (Olsen P) and (b) ammonium lactate–extractable phosphorus (P-AL).

 


View larger version (17K):
[in this window]
[in a new window]
  		Fig. 3. Effect of maximum phosphorus sorption capacity (PSCmax) on relationship between potential P release and soil test phosphorus (STP). The term * indicates significance at the 0.05 probability level, while ** indicates significance at the 0.01 probability level.

 


View larger version (14K):
[in this window]
[in a new window]
  		Fig. 4. Changes in relative content of CaCl2–extractable unreactive phosphorus (CaCl2–UP) promoted by CaCl2–extractable total phosphorus (CaCl2–TP) levels.

 
The DPS-1 estimate, calculated using Pox, Feox, and Alox according to Eq. [2], ranged from 7.7 to 16.8% in Level A and from 9.9 to 33.7% in the other levels (Table 4) and was poorly related to CaCl2–TP (r2 = 0.68). The values seem to be quite high for Level A, which was not fertilized with P since the start of the field experiment and appeared to be depleted. The method was developed for acid sandy soils in Central Europe while Swedish soils are young and contain acid soluble P such as apatite (Kirchmann and Eriksson, 1993; Kirchmann et al., 1996, 1999; Eriksson et al., 1997). The acid oxalate solution can dissolve apatite via formation of strong Ca–chelate complexes, rendering P soluble. In this way, oxalate can overestimate sorbed P in comparison with values obtained in more weathered acid sandy soils (Hartikainen, 1979). In two recent papers, Uusitalo and Tuhkanen (2000) and Peltovuori et al. (2002) found that ammonium oxalate overestimated labile P in Finnish soils, which also are young and rich in Ca-phosphate. They suggested sequential extraction of P with ammonium chloride (NH4Cl), ammonium fluoride (NH4F), and sodium hydroxide (NaOH) instead of Pox to estimate sorbed P. However, this is a laborious procedure and it is not suitable for using in risk assessment systems. So, alternative methods must be tested for estimation of DPS in young noncalcareous soils.

The DPS-2 and DPS-3 estimates were determined according to Eq. [1] using either Olsen P or P-AL (recalculated in mmol kg–1) to estimate sorbed P while PSCmax (in mmol kg–1) represented PSC (Table 4). Since PSCmax can be influenced by P fertilization (Bache and Williams, 1971), it was determined using soils from Level A, which had not received P fertilization since the start of the field experiment, and PSCmax was considered to represent maximum PSC in the soils. Like ammonium oxalate, the acid ammonium lactate solution can extract some Ca-bound phosphate. However, the complexes formed by the lactate and the acetate ions in the ammonium lactate solution are much weaker than the ones formed by the oxalic ions. The P-AL was approximately twice the amount of Olsen P. The result was that DPS values calculated with P-AL were approximately double those calculated with Olsen P. It should be kept in mind that comparison with the critical value of 25% that is usually used for DPS-1 (Breeuwsma and Silva, 1992) is not relevant, since the methods of analyses are not comparable. The CaCl2–TP was exponentially correlated to DPS-2 and DPS-3 (r2 ≥ 0.83; Table 4). For DPS-4 and DPS-5, PSC was determined according to Eq. [3]. Correlations with CaCl2–TP gave similar results as for DPS-2 and DPS-3 with r2 ≥ 0.79 (Table 4). This means that by considering PSC in addition to STP, soils with varying soil properties, such as clay content, pH, and PSC, can be compared regarding P release. The exponential relationship indicates that the strength of P bonding decreases as soils become more saturated with P (Barrow, 1978).

The ratios of Olsen P or P-AL (recalculated in mmol kg–1) to PSI (mmol kg–1) were also calculated. The PSI was determined in all P levels of the soils. As Levels B, C, and D were fertilized with P, sorbed P cannot be considered negligible compared with P sorption measured in Level A, and can thus affect PSI (Bache and Williams, 1971). As expected PSI decreased from Levels A to D but differences were not statistically significant (Table 3). Linear correlations were obtained when CaCl2–TP was plotted against the ratios of Olsen P to PSI and P-AL to PSI for all soils (r2 ≥ 0.86, significant at the 0.001 probability level; Fig. 5a, 5b) . Since CaCl2–RP constitutes the main part of TP at high soil P levels, ratios of Olsen P to PSI and P-AL to PSI were also related to CaCl2–RP (Fig. 5c, 5d) and similar linear correlations were obtained (r2 ≥ 0.90, significant at the 0.001 probability level).



View larger version (19K):
[in this window]
[in a new window]
  		Fig. 5. Relationship between CaCl2–extractable total phosphorus (CaCl2–TP) or CaCl2–extractable reactive phosphorus (CaCl2–RP) and the ratio of NaHCO3–extractable phosphorus (Olsen P) to phosphorus sorption index (PSI) or ammonium lactate–extractable phosphorus (P-AL) to PSI. The term *** indicates significance at the 0.001 probability level.

 
The results obtained in this study were similar to those reported by several other authors and summarized by Beauchemin and Simard (1999). However, most of the authors used DPS based on oxalate extraction. Beauchemin and Simard (1999) pointed out that the relationships were strong as long as the soils studied were fairly similar in texture and pH and that the relationships weakened as a wider range of soil types was included. However, in this study clay content ranged from 7 to 59% and pH ranged from 5.8 to 7.5, and still quite high correlations were obtained, indicating the possibility of using these methods on a wide range of soil types.

Both linear (Sharpley, 1995; Lookman et al., 1996) and nonlinear relationships (Kleinman et al., 2000) have been found between sorption saturation and P released in previous studies. When PSCmax or Feox and Alox were used to estimate PSC in this study, exponential relationships were obtained with CaCl2–TP, while linear relationships were obtained when PSI was used to estimate PSC. Values of PSCmax or Feox and Alox were not influenced by P fertilization and gave more or less constant denominators for each site resulting in exponential relationships, while PSI gave a slightly decreasing denominator with increasing soil P level, even if differences were not statistically significant, and resulted in linear relationships.

The DPS-2 and DPS-3 estimates are valuable for understanding the relationship between DPS and P release. However, PSCmax is laborious to obtain through sorption isotherms and normally, depleted soil samples such as those from Level A are not available for PSCmax analysis, so DPS-2 and DPS-3 are not suitable in risk assessment systems. Therefore, two estimates of PSC, PSCox and PSI, which are both well related to PSCmax as shown by Börling et al. (2001), were tested. The DPS-4 and DPS-5 estimates and the ratios of Olsen P to PSI and P-AL to PSI were well correlated to CaCl2–TP (r2 ≥ 0.79 and r2 ≥ 0.86, respectively). So, if values of Feox and Alox are available for soils, DPS-4 and DPS-5 could be a useful tool for predicting potential P release in risk assessment systems. However, since PSI is relatively easy to determine, and due to the high correlations both with CaCl2–TP and CaCl2–RP, the ratios Olsen P to PSI and P-AL to PSI seemed to be the most useful variables for predicting potential P release in risk assessment systems for Swedish noncalcareous soils.


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
To identify soils vulnerable to P losses, both transport mechanisms and P source need to be considered. When estimating the P source in a field, the potential P release of the soil is an important variable. In this study, potential P release was exponentially related to Olsen P and P-AL for individual soils. However, soils with high PSC released less P than soils with low PSC at the same STP level. Thus, STP alone cannot be used to estimate potential P release in risk assessment systems when soils with different PSC are being considered. Therefore, PSC of the soil needs to be taken into account when estimating potential P release.

In young soils, such as those used in this study, extraction with ammonium oxalate can dissolve apatite (Uusitalo and Tuhkanen, 2000) and may thus overestimate sorbed P. Therefore, DPS based on extraction with ammonium oxalate does not seem to be a good method for the soils in this study. The DPS-2 and DPS-3 estimates were well correlated (r2 ≥ 0.83) to CaCl2–TP. However, since PSCmax is laborious to estimate and unfertilized plots are rarely available, the two methods did not prove to be suitable for risk assessment systems. Since DPS-4 and DPS-5 were well correlated to CaCl2–TP (r2 ≥ 0.79) they could be useful tools for predicting potential P release in risk assessment systems, if values of Feox and Alox are available for soils. The ratios of Olsen P to PSI and P-AL to PSI resulted in good correlations with both CaCl2–TP and CaCl2–RP (r2 ≥ 0.86, significant at the 0.001 probability level). As those ratios are easily determined and well correlated with potential P release, their use in risk assessment systems for soils could be viable.


   ACKNOWLEDGMENTS
 
The study was conducted within the MISTRA FOOD 21 program. We are grateful to Prof. Thomas Kätterer and Dr. Lennart Mattsson for assistance in statistical analyses of the data. We are also grateful to Dr. Gerd Johansson for her contributions to the study.


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




This article has been cited by other articles:


Home page
 Soil Sci.Home page
R.K. Gupta, Yadvinder-Singh, J.K. Ladha, Bijay-Singh, J. Singh, G. Singh, and H. Pathak
Yield and Phosphorus Transformations in a Rice Wheat System with Crop Residue and Phosphorus Management
Soil Sci. Soc. Am. J., August 9, 2007; 71(5): 1500 - 1507.
[Abstract] [Full Text] [PDF]

Home page
 J. Environ. Qual.Home page
J. P. Casson, D. R. Bennett, S. C. Nolan, B. M. Olson, and G. R. Ontkean
Degree of Phosphorus Saturation Thresholds in Manure-Amended Soils of Alberta
J. Environ. Qual., October 27, 2006; 35(6): 2212 - 2221.
[Abstract] [Full Text] [PDF]

Home page
 Soil Sci.Home page
B. L. Allen and A. P. Mallarino
Relationships between Extractable Soil Phosphorus and Phosphorus Saturation after Long-Term Fertilizer or Manure Application
Soil Sci. Soc. Am. J., February 2, 2006; 70(2): 454 - 463.
[Abstract] [Full Text] [PDF]

Home page
 Soil Sci.Home page
H. Zhang, J. L. Schroder, R. L. Davis, J. J. Wang, M. E. Payton, W. E. Thomason, Y. Tang, and W. R. Raun
Phosphorus Loss in Runoff from Long-term Continuous Wheat Fertility Trials
Soil Sci. Soc. Am. J., December 2, 2005; 70(1): 163 - 171.
[Abstract] [Full Text] [PDF]

Home page
 J. Environ. Qual.Home page
D. V. Ige, O. O. Akinremi, and D. N. Flaten
Environmental Index for Estimating the Risk of Phosphorus Loss in Calcareous Soils of Manitoba
J. Environ. Qual., October 12, 2005; 34(6): 1944 - 1951.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (15)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Börling, K.
Right arrow Articles by Barberis, E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Börling, K.
Right arrow Articles by Barberis, E.
Agricola
Right arrow Articles by Börling, K.
Right arrow Articles by Barberis, E.
Related Collections
Right arrow Batch Studies
Right arrow Sorption/Exchange
Right arrow Ecological Risk Assessment
Right arrow Phosphorus
Right arrow Soil Analysis

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