Published online 5 July 2005
Published in J Environ Qual 34:1446-1450 (2005)
DOI: 10.2134/jeq2005.0028
© 2005 American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
SHORT COMMUNICATIONS
Disturbance of Water-Extractable Phosphorus Determination by Colloidal Particles in a Heavy Clay Soil from the Netherlands
G. F. Koopmansa,b,*,
W. J. Chardona and
C. van der Salma
a Alterra, Wageningen University and Research Centre (WUR), P.O. Box 47, 6700 AA, Wageningen, the Netherlands
b Department of Soil Quality, Wageningen University, WUR, P.O. Box 85, 6700 EC, Wageningen, the Netherlands
* Corresponding author (gerwin.koopmans{at}wur.nl)
Received for publication January 25, 2005.
 |
ABSTRACT
|
|---|
Water extraction methods are widely used to extract phosphorus (P) from soils for both agronomic and environmental purposes. Both the presence of soil colloids in soil water filtrates, and the contribution of colloidal P to the molybdate-reactive phosphorus (MRP) concentration measured in these filtrates, are well documented. However, relatively little attention has been given to the direct disturbance by colloids of MRP measurement. The objective of this paper is to show this influence found for water extracts with a soil to solution ratio of 1:60 (v/v) (Pw), obtained from a heavy clay soil in the Netherlands. Colloidal particles, which passed a 0.45-µm filter, caused a large overestimation of MRP. The low ionic strength of the Pw filtrates (on average 0.64 mmolc L–1) probably caused soil dispersion and increased detachment of colloids from soil during extraction. After NaCl addition, followed by 0.45-µm filtration, MRP was on average 93% lower. This can be ascribed to flocculation of colloids and removal by filtration. A low ionic strength can thus lead to the direct disturbance by colloidal particles of MRP measurement in waters from soils sensitive to release of colloids.
Abbreviations: MRP, molybdate-reactive phosphorus • Pw, water-extractable phosphorus at a soil to solution ratio of 1:60 (v/v)
|
INTRODUCTION
|
|---|
THE AVAILABILITY of phosphorus in soil for plant uptake has been studied extensively, and numerous extraction methods for the determination of available P have been developed. (Chardon et al., 1996). Water-extractable P is a widely used method for P fertilizer recommendations in Austria, Belgium, the Netherlands, and Switzerland (Ehlert et al., 2003). In the Netherlands, Pw (Sissingh, 1971) and water-extractable P at a soil to solution ratio of 1:2 (w/v) (Sonneveld et al., 1990) are used for P fertilizer recommendations for arable land and soils from horticulture, respectively. Extraction methods for determining the plant availability of P in agricultural soils are also used for the identification of soils exhibiting a high risk for P loss to surface waters. In many studies, strong relationships have been demonstrated between water-extractable P and P in surface runoff, furrow irrigation water, and leachate (Pote et al., 1996; McDowell and Sharpley, 2001a; Westermann et al., 2001; Maguire and Sims, 2002). Furthermore, Kleinman et al. (2002a)(2002b) showed strong relationships between P extractable from different animal manures by water and P loss from soil by surface runoff. Thus, water-extractable P is used for both environmental and agronomic purposes. Accurate measurement of the P concentration in water extracts is a prerequisite for assessing the availability of P for plant uptake or the risk of P transfer to surface waters. The necessity of analyzing a clear filtrate was mentioned by Kuo (1996), without discussing the factors causing a disturbance when the filtrate is not clear. For obtaining a clear extract, Kuo (1996) mentions the use of 0.01 M CaCl2 as an alternative for water (Schofield, 1955; Houba et al., 1986), although this often extracts much less P from soils than water (Olsen and Watanabe, 1970; Koopmans et al., 2001). With soils having a very low availability of P (e.g., tropical soils), the use of 0.01 M CaCl2 can lead to analytical problems (Schofield, 1955). A standard procedure is to filter water samples through a 0.45-µm filter to separate particulate P from dissolved P, followed by the measurement of the P concentration by the colorimetric acid molybdate method of Murphy and Riley (1962). The measured P concentration is often assumed to represent inorganic P. However, solubilization of colloidal-bound P by the acid molybdate reagent of Murphy and Riley (1962) in 0.45-µm filtrates obtained from soil waters such as soil solution, water extracts, and runoff water can cause an overestimation of the inorganic P concentration (Haygarth et al., 1997; Sinaj et al., 1998; Hens and Merckx, 2002; Turner et al., 2004). Colloids are operationally defined as particles between 1 nm and 1 µm in size and play an important role in the movement of reactive solutes through the soil (Kretzschmar et al., 1999). The effect of absorbance of light by soil colloids in water extracts on the measurement of MRP concentrations has not been established yet. The objective of this paper is to show results of a study on this topic using samples from a heavy clay soil.
|
MATERIALS AND METHODS
|
|---|
Soils
Samples were taken from a heavy clay soil under grassland of a dairy farm in Waardenburg, the Netherlands. The soil can be classified as a mesic Ustic Endoaquert (Soil Survey Staff, 1996). The field has a size of approximately 90 by 175 m. In the past few years, the field received 87 kg P ha–1 yr–1, with 64 and 22% of the added P being from the application of cattle manure slurry and P fertilizer, respectively, and the remainder being from dung deposited by grazing cattle. The sampling protocol used was designed to obtain information on P losses to ground and surface waters within the framework of another study. Samples were taken from four different horizons (A, Cg, Cgr, and Cr) to a depth of 150 cm below soil surface, at five positions chosen randomly along four transects of the field. Soil was dried at 40°C and broken to pass a 2-mm sieve. For the analysis of CaCO3 and the particle size distribution, samples were bulked per horizon (n = 4). For the analysis of organic matter, pH, and the different P extractions including the experiments with Pw (see below), soils were bulked per horizon for each transect (n = 16). For Pw, additional samples were taken from the 0- to 10-cm soil layer at twenty positions chosen randomly along the four transects; after pretreatment, these were bulked per transect (n = 4).
Soil Analyses
The pH (H2O), organic matter, and CaCO3 were determined using standard analytical procedures (Houba et al., 1995). The pH was measured in a settling 1:5 (w/v) suspension of soil in water. Organic matter was determined by loss-on-ignition; a correction was made for water bound to the mineral phase. Calcium carbonate was determined by measuring the volume of CO2 produced after the addition of HCl. Particle size distribution was determined via pipette and sieve after removal of CaCO3 and organic matter (Houba et al., 1995). Inorganic P was determined after extraction of soil with 0.5 M H2SO4, and total P after combustion of organic matter in a muffle furnace (550°C), followed by extraction with 0.5 M H2SO4 (Kuo, 1996). Organic P was calculated as the difference between total and inorganic P. As in all other extracts, concentration of MRP was measured colorimetrically at 882 nm (Murphy and Riley, 1962). Phosphorus in 1:10 (w/v) 0.01 M CaCl2 extracts was determined according to Houba et al. (1986). The CaCl2 extracts were shaken on a reciprocating shaker at 165 strikes per minute. After centrifugation (1800 x g), MRP was measured. The Pw was determined according to Sissingh (1971). In brief, after 22 h of pre-equilibrating 1.2 mL of soil with 2 mL of water, 70 mL of water was added and the soil suspension was shaken end-over-end at 30 rounds per min for 1 h, followed by filtration over a Schleicher & Schuell (Dassel, Germany) 602H filter (0.45 µm). To obtain information on the potential direct disturbance by colloids during MRP measurement, we divided the Pw filtrate into three subsamples. In the first subsample MRP was measured. In the second subsample, turbidity was measured at 882 nm without addition of the molybdate reagent. Instead of the reagent, water was added to dilute this subsample. To the third 25-mL subsample, 0.75 g NaCl was added, corresponding with 0.51 M NaCl, as recommended by Sissingh (1971). After settlement of colloidal particles, the subsample was again filtered over a 0.45-µm filter and MRP was measured. The addition of NaCl does not affect the measurement of MRP (Sissingh, 1971). In the Netherlands, the Pw is normally reported in mg P2O5 L–1 of soil but in this study in mg P kg–1. Electrical conductivity was measured in the Pw filtrate, before and after NaCl addition and refiltration. Ionic strength was estimated from the measured electrical conductivity according to Griffin and Jurinak (1973).
Statistical Analyses
The least significant difference method, with a probability value of 0.05, was used to determine significant differences for the means of Pw and ionic strength between the four soil horizons (A, Cg, Cgr, and Cr), and before and after NaCl addition and refiltration. Results from the 0- to 10-cm soil layer were excluded from the statistical analysis because they were aliased with those from the A horizon. Statistical analysis was performed with Genstat 5, Release 7.1 (Genstat, 2003). Linear regression analysis was done using the same software package. Significance of R2adj values was determined using F tests.
|
RESULTS AND DISCUSSION
|
|---|
Soils
The physical and chemical characteristics of the soils used in this study are presented in Table 1. The pH and CaCO3 content increased with depth while the opposite was true for the organic matter content. Total P amounted to 747 mg P kg–1 in the A horizon and was larger than in the other horizons. Organic P amounted to 451 mg P kg–1 and showed a clear decrease with depth.
Water-Extractable Phosphorus
The values measured for Pw were much higher than expected, because the heavy clay soil used in our study is known to have a large P-fixing capacity (Reijneveld, 2001). Therefore, we tested if these high Pw values were the result of the presence of colloids in the Pw filtrates. Absorbance of light at 882 nm was measured in subsamples of the Pw filtrates to which no molybdate reagent of Murphy and Riley (1962) was added. Figure 1
shows the strong linear correlation 
between the absorption of subsamples without the molybdate reagent and the Pw measured in subsamples with the reagent. When we plotted absorbance instead of the calculated Pw value measured in the subsamples with the molybdate reagent, data points closely followed the 1:1 line (not shown). Thus, our results can be explained by the absorbance of light by colloids with a size smaller than 0.45 µm present in the Pw filtrates. Although all samples were taken from the same field, a large variation in turbidity was found in the water extracts (Fig. 1). This can probably be ascribed to variation in clay, CaCO3, and organic matter contents between soil horizons (Table 1). Subsamples of the Pw filtrates from the Cr horizon had the lowest turbidity (Fig. 1). The soil samples from this horizon had, on average, the lowest clay and highest CaCO3 contents (Table 1). Furthermore, the ionic strength of the Pw filtrates before NaCl addition obtained from the Cr horizon was significantly higher than for the other soil horizons, possibly leading to less dispersion of colloids (Table 2). After addition of NaCl to subsamples of the same Pw filtrates, followed by 0.45-µm filtration, MRP concentrations were significantly lower; Pw decreased from an overall average of 11.4 to 0.7 mg P kg–1 (Fig. 1, Table 2). However, no significant effect of soil horizon was found for Pw. The much lower Pw can be explained by the removal of colloidal particles from the Pw filtrates by increasing the ionic strength due to NaCl addition, causing coagulation of colloids, which were then retained by the 0.45-µm filter. Addition of NaCl increased ionic strength significantly, from an overall average of 0.64 to 526 mmolc L–1 (Table 2). The Pw measured after the addition of NaCl shows a linear relationship with P extracted by 0.01 M CaCl2 (R2adj = 69.1%) (Fig. 2)
, whereas there was no relationship between the Pw measured before the addition of NaCl and 0.01 M CaCl2–extractable P (not shown).
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 1. Relationship between water-extractable phosphorus (Pw) measured after addition of the molybdate reagent, and absorption measured at 882 nm in a second subsample of the same filtrate without the molybdate reagent. The filled squares represent the Pw values measured in a third subsample, after addition of NaCl and 0.45-µm filtration. The symbol *** indicates significance at the 0.001 probability level.
|
|
View this table:
[in this window]
[in a new window]
|
Table 2. Measured values of water-extractable phosphorus (Pw) and ionic strength before and after NaCl addition and refiltration.
|
|
View larger version (13K):
[in this window]
[in a new window]
|
Fig. 2. Relationship between P extractable with 1:10 (w/v) 0.01 M CaCl2, and water-extractable phosphorus (Pw) measured after addition of NaCl and 0.45-µm filtration (open squares). The filled square deviates significantly from the relationship. The symbol *** indicates significance at the 0.001 probability level.
|
|
The presence of colloidal particles in 0.45-µm filtrates obtained from river waters, soil solution, surface runoff, and soil water extracts is well documented (e.g., Haygarth et al., 1997). These soil water extracts were obtained at varying soil to solution ratios (all w/v): 1:1 (Hens and Merckx, 2002), 1:4 (Dolfing et al., 1999), 1:10 (Sinaj et al., 1998; Hens and Merckx, 2002), 1:30 (Heathwaite et al., 2005), and 1:200 (Turner et al., 2004). Thus, the presence of colloids in water extracts is not limited to wide soil to solution ratios. As mentioned in the introduction section, solubilization of colloidal-bound P by the acid molybdate reagent can contribute to the measured MRP concentration. In water extracts from calcareous soils, dissolution of colloidal Ca-P minerals in the acid molybdate reagent can also contribute to MRP (Sinaj et al., 1998; Turner et al., 2004). However, the direct influence of soil colloids on MRP measurement by light absorbance, as found in our study, has received much less attention. A low ionic strength of soil waters causes dispersion of soil and, consequently, increases detachment of colloids from soil (Kretzschmar et al., 1999). This has probably caused the presence of colloidal particles in our Pw filtrates, because the ionic strength of these filtrates before NaCl addition was low (Table 2). It was on average a factor 50 lower than the ionic strength of a 0.01 M CaCl2 extract, which frequently has been used as a surrogate for soil solution (Houba et al., 1986). Also in other studies, a low ionic strength was found when determining water-extractable P in soil using a wide soil to solution ratio (e.g., Yli-Halla et al., 1995; Yli-Halla and Hartikainen, 1996). In the study of Yli-Halla et al. (1995), a low ionic strength was also found in surface runoff, which can be explained by the fact that in runoff, a low amount of soil comes into contact with a large volume of solution. Therefore, in soils sensitive to detachment of colloids (e.g., clay soils and soils with a low aggregate stability), there is a risk of direct disturbance by colloidal particles of MRP measurement in solution.
The results of our study are important, because soil water extracts are widely used for assessing the availability of P for plant uptake and the risk of P transfer to surface waters. To reduce the direct disturbance by colloids, ion chromatography can be used to obtain a true measurement of the ortho-P concentration. This has also the advantage of reducing the contribution of acid-mediated solubilization of colloidal P (McDowell and Sharpley, 2001b). However, during the measurement of P using ion chromatography, P can still adsorb to colloids, causing an underestimation of the ortho-P concentration (Sinaj et al., 1998). Furthermore, ion chromatography has currently a relatively high detection limit (McDowell and Sharpley, 2001b). High-speed centrifugation was suggested by Kuo (1996) for obtaining a clear water extract. To reduce the influence of colloids on measurement of MRP, Sinaj et al. (1998) proposed to use 0.2- or 0.025-µm filters; however, clogging of these fine filters is a problem, reducing the effective filter size. Furthermore, Turner et al. (2004) demonstrated the presence of colloidal Ca-P minerals with a size ranging between 3 and 0.3 nm in water extracts obtained from calcareous soils. For turbid water filtrates, we therefore propose to flocculate colloidal particles by adding a salt like NaCl [0.75 g NaCl 25 mL–1 water filtrate (Sissingh, 1971), proved to be effective for our samples], followed by filtering through a standard filter (0.45 µm) and MRP measurement. However, other salts could also be tested for flocculation.
CONCLUSIONS
Colloidal particles with a size smaller than 0.45 µm caused an overestimation of the MRP concentrations measured in water extracts obtained at a wide soil to solution ratio of 1:60 (v/v). After addition of NaCl, followed by 0.45-µm filtration, MRP concentrations were much lower. The low ionic strength of the Pw filtrates (on average 0.64 mmolc L–1) probably caused soil dispersion and increased detachment of colloids from soil during extraction. Under the condition of a low ionic strength, there is thus a risk of direct disturbance of MRP measurement by colloidal particles in soil waters (e.g., soil solution, water extracts, and runoff water) from soils sensitive to detachment of colloids. In this case, the availability of P for plant uptake or the risks of P transfer to surface waters are strongly overestimated.
|
ACKNOWLEDGMENTS
|
|---|
The authors thank René Aalderink and Antonie van den Toorn for collecting the soil samples and Oene Oenema and Paul Römkens for their valuable comments on a previous version of this manuscript.
|
REFERENCES
|
|---|
- Chardon, W.J., R.G. Menon, and S.H. Chien. 1996. Iron oxide impregnated filter paper (Pi test): A review of its development and methodological research. Nutr. Cycling Agroecosyst. 46:41–51.
- Dolfing, J., W.J. Chardon, and J. Japenga. 1999. Association between colloidal iron, aluminum, phosphorus, and humic acids. Soil Sci. 164:171–179.[CrossRef]
- Ehlert, P.A.I., C. Morel, M. Fotyma, and J. Destain. 2003. Potential role of phosphate buffering capacity of soils in fertilizer management strategies fitted to environmental goals. J. Plant Nutr. Soil Sci. 166:409–415.[CrossRef]
- GenStat. 2003. GenStat statistical package for Windows. Lawes Agric. Trust, Rothamsted Exp. Stn., UK.
- Griffin, R.A., and J.J. Jurinak. 1973. Estimation of activity coefficients from the electrical conductivity of natural aquatic systems and soil extracts. Soil Sci. 116:26–30.
- Haygarth, P.M., M.S. Warwick, and W.A. House. 1997. Size distribution of colloidal molybdate reactive phosphorus in river waters and soil solution. Water Res. 31:439–448.[CrossRef]
- Heathwaite, A.L., P.M. Haygarth, R. Matthews, N. Preedy, and P. Butler. 2005. Evaluating colloidal phosphorus delivery to surface waters from diffuse agricultural sources. J. Environ. Qual. 34:287–298.[Abstract/Free Full Text]
- Hens, M., and R. Merckx. 2002. The role of colloidal particles in the speciation and analysis of "dissolved" phosphorus. Water Res. 36:1483–1492.[Medline]
- Houba, V.J.G., I. Novozamsky, A.W.M. Huybregts, and J.J. van der Lee. 1986. Comparison of soil extractions by 0.01 M CaCl2, by EUF and by some conventional extraction procedures. Plant Soil 96:433–437.
- Houba, V.J.G., J.J. van der Lee, and I. Novozamsky. 1995. Soil analysis procedures; Other procedures (soil and plant analysis, Part 5B). Agric. Univ., Wageningen, the Netherlands.
- Kleinman, P.J.A., A.N. Sharpley, B.G. Moyer, and G.F. Elwinger. 2002a. Effect of mineral and manure phosphorus sources on runoff phosphorus. J. Environ. Qual. 31:2026–2033.[Abstract/Free Full Text]
- Kleinman, P.J.A., A.N. Sharpley, A.M. Wolf, D.B. Beegle, and P.A. Moore. 2002b. Measuring water-extractable phosphorus in manure as an indicator of phosphorus in runoff. Soil Sci. Soc. Am. J. 66:2009–2015.[Abstract/Free Full Text]
- Koopmans, G.F., M.E. van der Zeeuw, P.F.A.M. Römkens, W.J. Chardon, and O. Oenema. 2001. Identification and characterization of phosphorus-rich sandy soils. Neth. J. Agric. Sci. 49:369–384.
- Kretzschmar, R., M. Borkovec, D. Grolimund, and M. Elimelech. 1999. Mobile subsurface colloids and their role in contaminant transport. Adv. Agron. 66:121–193.
- Kuo, S. 1996. Phosphorus. p. 869–919. In D.L. Sparks (ed.) Methods of soil analysis. Part 3. Chemical methods. SSSA Book Ser. 5. SSSA, Madison, WI.
- Maguire, R.O., and J.T. Sims. 2002. Soil testing to predict phosphorus leaching. J. Environ. Qual. 31:1601–1609.[Abstract/Free Full Text]
- McDowell, R.W., and A.N. Sharpley. 2001a. Approximating phosphorus release from soils to surface runoff and subsurface drainage. J. Environ. Qual. 30:508–520.[Abstract/Free Full Text]
- McDowell, R.W., and A.N. Sharpley. 2001b. Soil phosphorus fractions in solution: Influence of fertiliser and manure, filtration and method of determination. Chemosphere 45:737–748.[Medline]
- Murphy, J., and J.P. Riley. 1962. A modified single solution method for the determination of phosphate in natural waters. Anal. Chim. Acta 27:31–36.[CrossRef]
- Olsen, S.R., and F.S. Watanabe. 1970. Diffusive supply of phosphorus in relation to soil textural variations. Soil Sci. 110:318–327.
- Pote, D.H., T.C. Daniel, A.N. Sharpley, P.A. Moore, Jr., D.R. Edwards, and D.J. Nichols. 1996. Relating extractable soil phosphorus to phosphorus losses in runoff. Soil Sci. Soc. Am. J. 60:855–859.[Abstract/Free Full Text]
- Reijneveld, J.A. 2001. Effects of decreased phosphorus addition on phosphate fixing clay soils. (In Dutch.) Rep. 27. Plant Res. Int., Wageningen, the Netherlands.
- Schofield, R.K. 1955. Can a precise meaning be given to "available" soil phosphorus? Soils Fert. 43:373–375.
- Sinaj, S., F. Mächler, E. Frossard, C. Faïsse, A. Oberson, and C. Morel. 1998. Interference of colloidal particles in the determination of orthophosphate concentrations in soil water extracts. Commun. Soil Sci. Plant Anal. 29:1091–1105.
- Sissingh, H.A. 1971. Analytical technique of the Pw method, used for the assessment of the phosphate status of arable soils in the Netherlands. Plant Soil 34:483–486.[CrossRef]
- Soil Survey Staff. 1996. Keys to soil taxonomy. 7th ed. USDA-NRCS, Washington, DC.
- Sonneveld, C., J. van den Ende, and S.S. de Bes. 1990. Estimating the chemical compositions of soil solutions by obtaining saturation extracts or specific 1:2 by volume extracts. Plant Soil 122:169–175.
- Turner, B.L., M.A. Kay, and D.T. Westermann. 2004. Colloidal phosphorus in surface runoff and water extracts from semiarid soils of the western United States. J. Environ. Qual. 33:1464–1472.[Abstract/Free Full Text]
- Westermann, D.T., D.L. Bjorneberg, J.K. Aase, and C.W. Robbins. 2001. Phosphorus losses in furrow irrigation runoff. J. Environ. Qual. 30:1009–1015.[Abstract/Free Full Text]
- Yli-Halla, M., and H. Hartikainen. 1996. Release of soil phosphorus during runoff as affected by ionic strength and temperature. Agric. Food Sci. Finl. 5:193–202.
- Yli-Halla, M., H. Hartikainen, P. Ekholm, E. Turtola, M. Puustinen, and K. Kallio. 1995. Assessment of soluble phosphorus load in surface runoff by soil analyses. Agric. Ecosyst. Environ. 56:53–62.[CrossRef]
TOP
ABSTRACT
INTRODUCTION
