Soil phosphorus sorption and availability as a function of high phosphorus fertilizer additions

Abstract

After a 3‐months equilibration of soil with phosphorus (P) (up to four times the respective P sorption capacity), equilibrium P concentration (EPCo), standard P requirement (SPR), P sorption index (SI), and P availability by Bray I, Olsen, water and iron‐oxide paper strip methods were determined on three soils of the Latium region of Italy, widely ranging in their affinity for P. Soil P addition increased EPCo and availability P content and decreased SPR and SI values for all soils with differences between soil types a ffinction of P sorption maximum. The tractional increase of available NaHCO₃‐P with added P, i.e. P availability index (F) was 0.486 for the soil with the lowest P sorption maximum, 0.217 for the soil with the highest P sorption maximum, and 0.369 for the third soil presenting an intermediate P sorption (r = 0.997; P<0.01). The results indicate that soil type, in addition to the amount of P added, will determine the potential for a soil to release P to runoff.     

Changes in soluble and equilibrium phosphate concentration in selected soils from Italy

Abstract

Soil phosphate in solution (PsolCpand equilibrium P concentration (EPCo) are important soil P parameters both for agronomical and environmental purposes. Solution P is the main source from which plant roots adsorb P, whereas EPCo give information about the amount and direction of changes between soluble and particulate P that occur during transport of sediment in stream flow. Changes in Psol and EPCo with soil extractable P (Mehlich 3‐P and Olsen P) were determined following equilibration of 13 Italian soils with five rates of fertilizer P (0 to 100 mg kg^‐1). The slope of the regression equation (Y=a+bX) for the relationships between Psol and EPCo (Y), and soil extractable P (X) decreased with increasing soil P sorption. Furthermore, at the same level of Mehlich 3 and Olsen P, higher sorbing P soils had lower values of Psol and EPCo. As a result, changes in Psol and EPCo per unit increase of soil extractable P were closely related (R² of 0.86 to 0.93) to soil P Sorption Index (SI). Equilibrium P Concentration could be estimated from soil solution P for the studied soils.     

The effect of soil phosphorus on particulate phosphorus in land runoff

Accumulation of surplus phosphorus (P) in the soil and the resulting increased transport of P in land runoff contribute to freshwater eutrophication. The effects of increasing soil P (19–194 mg Olsen‐P (OP) kg⁻¹) on the concentrations of particulate P (PP), and sorption properties (Q_max, k and EPCo) of suspended solids (SS) in overland flow from 15 unreplicated field plots established on a dispersive arable soil were measured over three monitoring periods under natural rainfall. Concentrations of PP in plot runoff increased linearly at a rate of 2.6 μg litre⁻¹ per mg OP kg⁻¹ of soil, but this rate was approximately 50% of the rate of increase in dissolved P (< 0.45 μm). Concentrations of SS in runoff were similar across all plots and contained a greater P sorption capacity (mean + 57%) than the soil because of enrichment with fine silt and clay (0.45–20 μm). As soil P increased, the P enrichment ratio of the SS declined exponentially, and the values of P saturation (P_sat; 15–42%) and equilibrium P concentration (EPCo; 0.7–5.5 mg litre⁻¹) in the SS fell within narrower ranges compared with the soils (6–74% and 0.1–10 mg litre⁻¹, respectively). When OP was < 100 mg kg⁻¹, P_sat and EPCo values in the SS were smaller than those in the soil and vice‐versa, suggesting that eroding particles from soils with both average and high P fertility would release P on entering the local (Rosemaund) stream. Increasing soil OP from average to high P fertility increased the P content of the SS by approximately 10%, but had no significant (P > 0.05) effect on the P_sat, or EPCo, of the SS. Management options to reduce soil P status as a means of reducing P losses in land runoff and minimizing eutrophication risk may therefore have more limited effect than is currently assumed in catchment management.     

Addition of phosphorus to soils with low to medium phosphorus retention capacities. II. effect on soil phosphorus extractability

Abstract

Knowledge of the change in soil extractable phosphorus (P) as a consequence of soil P fertilization could be useful in discriminating soils with a potential for soil P release to runoff or movement of P along the soil profile. In this research, soils with low to medium P retention capacity were equilibrated for 90 days with soluble P (KH₂PO₄) at rate of 100 mg P kg^‐1 soil. After this period, soil samples both with and without the P addition were analyzed using six conventional methods: 1) Olsen, 2) Bray 1,3) Mehlich3,4) Egner, 5) Houba, dilute CaCl₂ solution, and 6) distilled water, and three “innovative”; P‐sink methodologies: 1) Fe oxide‐coated paper strip, 2) anion exchange resin membrane, and 3) cation‐anion exchange resin membrane. The soils without P addition had low levels of extracted P as determined by all nine procedures. Net increases in the amount of P extracted from the soils with added P ranged from 4.2 mg kg^‐1 (CaCl₂ extraction) to 57.6 mg kg^‐1 (cation‐anion resin membrane extraction). Relationships between change in extracted P and i) physical and chemical characteristics, and ii) soil P sorption properties are also presented and discussed.     

Release of soil phosphate by sequential extractions as a function of soil properties and added phosphorus

Abstract

A study of sequential phosphate (P) extraction by water and iron oxide‐impregnated paper strip procedures was carried out on three Italian soils ranging widely in soil characteristics and enriched with three rates of fertilizer P. The degree of change was dependant on P addition, soil P properties, and type of extraction. For the Fe‐oxide strip procedure, a greater release of P than for water extraction was observed for soils with and without added P. At a given level of added P, more P was released from the soil with the lowest P sorption index (SI). However, at a given level of NaHCO₃‐extractable P, less P was released from the soil with lower SI than from soil with a higher SI, indicating that a greater available P content was necessary for low P sorbing soils to maintain a given rate of P release. The variation of SI accounted for 96% and 92% of the variation in amount of water‐extractable and Fe‐oxide strip P at a given P addition. Furthermore, SI accounted for 97% and 98% of the variation in water‐extractable and Fe‐oxide P at a given increase in available soil P. Inclusion in a soil testing program of an estimate of the P Sorption Index, that accounts for the overall effect of soil properties affecting sorption in soils (clay content and type, iron and aluminum oxide content, surface area, etc.), may improve fertilizer P requirements for optimum crop growth for certain soils.     

Changes in some soil phosphorus availability parameters as induced by phosphorus addition and soil sorption properties

Abstract

Changes in agronomic and environmental soil phosphorus (P) availability parameters, i.e., Mehlich‐ and Olsen‐extractable P, reversibly‐adsorbed P, soil‐solution P, and equilibrium‐P concentration were determined following equilibration of 13 Italian soils with five rates of P application (0, 12.5, 25, 50, and 100 mg P kg^‐1 soil). Soil P availability as determined by each parameter increased with added P. The relative change in soil P availability with added P was a function of soil sorption index silicon (SI), according to the equation DP=(P_added)^a*exp(b+g*SI). This equation accounted for 94 to 98% of the variance in soil‐P availability. The inclusion of SI in a soil testing program may increase the reliability in assessing both soil‐P fertilizer requirements and the vulnerability of a soil to P loss in runoff following land application of fertilizer or manure P.     

Prediction of phosphorus requirements of Desmodium intortum cv. greenleaf using a phosphorus sorption index and extractable phosphorus

Abstract

The rates of applied phosphorus required for 90% maximum yield of Desmodiim intortum cv. Greenleaf were calculated from pot experiments using 24 fertilized and unfertilized soils from the Atherton Tableland, Queensland, Australia.

Phosphorus required was highly correlated (r² = 0.94) with the phosphorus sorbed (P sorbed) by the soils at a supernatant solution P concentration of 0.08 ppm. P sorbed was found to be a function of phosphorus buffer capacity at 0.08 ppm ("PBC") and phosphorus extractable by acid (0.005 M H₂S0₄) or bicarbonate (0.5 M NaHCO₃). PBC was highly correlated (r² = O.84) with a phosphorus sorption index ("PSI") derived from one addition of 500 μg P g^‐1 soil.

Combining PSI with acid or bicarbonate extractable P in a multiple regression equation allowed the estimation of phosphorus required with multiple correlation coefficients of R² = 0.80 and R² = 0.83 respectively.     

Assessing alternative approaches to predicting soil phosphorus sorption

Twenty‐five pasture soils were sampled from high‐rainfall zones of southeastern Australia to examine relationships between soil properties, and between soil properties and P buffering capacity (PBC) measures. Correlations between PBC values and soil properties were generally poor, with the exception of oxalate‐extractable Al (Al_ox) (r ≥ 0.97). Predictions of PBC were further improved when clay, as well as Al_ox, was included in a linear regression model (r² ≥ 0.98). When Al_ox and oxalate‐extractable Fe were excluded from the modelling exercise, a more complex three‐term linear regression model, including pH_H2O, exchangeable H and cation exchange capacity, adequately fitted both PBC values of the 25 soils examined in this study (r² ≥ 0.76). However, the Al_ox, Al_ox plus clay and the three‐term models gave poor predictions of the PBC values when the models were validated using 28 independent soils. These results emphasize the importance of model validation, because predictive models based on soil properties were not robust when tested across a broader range of soil types. In comparison, direct measures of PBC, such as single‐point P sorption measures, are more practical and robust methods of estimating PBC for Australian soils.     

Sorption of phosphorus by some surface soils from Quebec in relation to their properties

Abstract

Surface horizons from Podzolic and Gleysolic soils were collected in various parts of the province of Quebec, Canada, and equilibrated with various amounts of KH₂PO₄ in 0.01 M CaCl₂ for 48 hours. P sorption data conformed to the linear form of the Langmuir and Freundlich equations. P solubility isotherms showed evidence of hydroxyapatite formation in most samples studied, whereas equilibration solutions of only few samples were saturated with respect to either dicalcium phoshate dihydrate or octocalcium phosphate. These reaction products were associated to soil pH and levels of added phosphate. The average values of the Langmuir sorption maximum for these studied Gleysolic and Podzolic samples were 763 and 1096 μg/g respectively. These values were higher than those obtained by the segmented and modified Freundlich models.

Relationships between the soil characteristics and P sorption parameters were evaluated by regression analysis. Among all variables, oxalate‐extractable Fe plus Al content of the Podzolic samples and the ratio of oxalate—extractable Al to clay of the Gleysolic samples gave the best significant correlation coefficients. Furthermore, soil pH and various ratios such as pyrophosphate‐extractable Fe and Al, oxalate‐extractable Fe and organic matter to clay were found to be significantly correlated only with the P sorption parameters of the Gleysolic samples.     

Economics of phosphorus fertilization of barley as influenced by concentration of extractable phosphorus in soil

Abstract

Field experiments were conducted at 60 sites in central and north‐central Alberta to determine the yield response of barley (Hordeum vulgare L.) to phosphorus (P) fertilizer and economics of P application on soils with different concentrations of extractable P in the 0–15 cm soil layer. On the unfertilized plots, barley yield increased with increasing concentration of extractable P in the soil up to 22 mg P kg^‐1, but the yield response to applied P decreased. The net present value (NPV) of returns from P fertilization increased with increasing rate of P up to approximately 51 kg P₂O₅ ha^‐1. The NPV of applied P decreased with increasing concentration of extractable P in soil. On soils with extractable P more than 22 mg P kg^‐1, P application did not result in positive NPV.     

Sequentially extracted phosphorus fractions in peat‐derived soils

Phosphorus (P) forms were sequentially extracted from peat derived soils (Eutric Histosols and Gleysols) at eight sites in Saxony‐Anhalt (Germany) to disclose general differences in P pools between mineral and organic soils and to investigate effects of peat humification and oxidation in conjunction with land use and soil management on the P status of soils. Overall 29 samples providing a wide variety of basic chemical properties were subjected to the Hedley fractionation. The Histosol topsoils contained more total P (P_t) (1345 ± 666 mg kg^—1) than the Gleysol topsoils (648 ± 237 mg kg^—1). The predominant extractable fractions were H₂SO₄‐P (36—63 % of P_t) in calcareous and NaOH‐P_o (0—46 % of P_t) in non‐calcareous Histosols. These soils had large pools of residual P (13—93 % of P_t). Larger contents and proportions of P_o and of labile P fractions generally distinguished organic from mineral soils. Regression analyses indicated that poorly crystalline pedogenic oxides and organic matter were binding partners for extractable and non‐extractable P. Intensive management that promotes peat humification and oxidation results in disproportional enrichments of labile P fractions (resin‐P, NaHCO₃‐P_i, and NaHCO₃‐P_o). These changes in P chemistry must be considered for a sustainable management of landscapes with Histosols and associated peat derived soils.     

Statistical evaluation of phosphorus breakthrough profiles using a segmented exponential model

Abstract

Few studies have utilized a statistical approach for the evaluation and comparison of breakthrough profiles obtained from soil column investigations. Our objective was to investigate the use of nonlinear estimation techniques to identify the phosphorus (P) breakthrough point and other parameters endemic to breakthrough data. Four soils exhibiting a range in physical and chemical properties were leached under saturated‐flow conditions with a 10 mg P/L solution for 34 d. Leachate P concentrations were analyzed by flow injection colorimetric analysis. A four‐parameter segmented exponential model provided a good fit for the breakthrough data in each soil tested. In addition to identifying the breakthrough point, the model parameters provided estimates of other identifiable aspects of a breakthrough profile including background P concentration, equilibrium P concentration, and the rate of concentration change once breakthrough is attained. Estimation of these parameters also facilitated statistical comparison of breakthrough profiles from different soils. Single and joint hypothesis tests showed that significant differences existed among the breakthrough profiles, the breakthrough points, and the rates of concentration change. Additionally, simulation was used to examine the correlation structure between the estimated breakthrough point and 17 soil properties. The simulation showed that 6 soil properties— ammonium oxalate extractable aluminum (Al) and iron (Fe), citrate dithionite extractable Al, CaCl₂ and acetate extractable P, and P sorption capacity— were strongly correlated with P breakthrough.     

Phosphorus sorption by four soils receiving long‐term applications of fertilizer

Abstract

A laboratory study was conducted to evaluate P sorption in the Ap horizon of four soil series in the Ultisol order (Benndale Is, Hartsells fsl, Lucedale fsl, and Dewey sicl) receiving the same fertility treatments since 1929. Soil was collected in the spring of 1985 from 4 treatments: i) no‐lime, plus P (total fertilizer P = 1584 kg/ha from 1929 to 1985); ii) no‐K, plus P (total fertilizer P = 1584 kg/ha); iii) low‐P (total fertilizer P = 442 kg/ha); 4) standard treatment (total fertilizer P = 2376 kg/ha). The soils and treatments within a soil varied in pH, total P, Mehlich 1 extractable P, K, Ca and Mg, and KC1 extractable Al. The four soils had large differences in P sorption capacity which increased with increasing clay content. The Dewey (27 % clay) soil had the highest P sorption capacity and the Benndale (4 % clay) soil had the smallest P sorption capacity. Sorption of P within a soil was affected by the rate of added P and past fertility treatment. Treatment differences in P sorption were due primarily to the level of extractable P and soil pH. Within a given soil, P sorption (at a given rate of added P) generally decreased as the level of extractable P increased. Regression analysis of P sorption data for equilibrium P concentrations of 1 to 32 μmol/L showed that the parti‐ tioning between sorbed and solution P (buffer power) had not been changed by 56 years of annual applications of P. The maximum P sorption capacity of the four soils was decreased slightly by P fertilization.     

Phosphorus‐sorption characteristics of intensely weathered soils in south‐central kentucky

Abstract

Studies were conducted to investigate phosphorus (P)‐sorption characteristics of some intensely weathered soils in south‐central Kentucky. Phosphorus adsorption characteristics reflected the chemical and mineralogical properties of the soils studied. All adsorption data were adequately described by first order kinetic reactions which implied that the soils have uniform surfaces for P sorption. In spite of the limitations of the Langmuir equation, its usefulness in summarizing data into one adsorption maximum value was demonstrated by nearly identical adsorption maxima estimated by three linear transformations of the equation and small deviations from the observed maxima. Variations in adsorption maxima between surface and subsoils and among soils were best correlated with extractable aluminum (Al) (r = 0.93, p<0.01) and crystalline iron (Fe) oxy‐hydroxides (r = 0.97, p<0.01). Clay content was also highly correlated with P sorption (r = 0.97, p <0.01) as well as with extractable Al (r = 0.83, p<0.05) and crystalline Fe oxides (r = 0.92, p<0.01) suggesting that its contribution may have been through its association with these soil components. In contrast, organic matter had a negative association with P sorption (r = ‐0.83, p<0.05). The results indicate higher P sorption in subsoil than in surface horizons, controlled mainly by extractable Al and crystalline Fe oxyhydroxides.     

Relationship between phosphorus sorption indexes and plant response in a greenhouse experiment

Abstract

Twenty six phosphorus (P) sorption indexes were utilized as complement measurements of P fertility diagnosis on sixteen soils from Sandy Pampa and Ondulating Pampa. Adsorption values behaved as non‐availability indexes and were negatively correlated with P absorption by plants. The best parameters were P‐25–2hrs: adsorbed P with 25 μg P added, equilibrium period: 2 hrs (r = ‐0.89**); BC: buffer capacity (r = 0.86**); P‐1550: adsorbed P with 1550 (μg P added (r = ‐0.80* in Ondulating Pampa). Intensity or desorption related positively with plant P (r = 0.69**).

When applying P fertilization in a greenhouse trial, absorbed P, dry matter, and foliar area showed a positive correlation with adsorption indexes and a negative one with desorption indexes.     

Phosphorus sorption and availability indices as affected by properties of calcareous soils

Abstract

The Olsen solution is usually considered the best extractant for estimating P availability in calcareous soils, but predictability of the response to P fertilizers is often low under field conditions. In this study, soil characteristics influencing P sorption and extractability were evaluated. Forty‐one soils varying in CaCO₃, pH, and clay content were selected from pastures to minimize the effect of recent P additions. A P sorption index (PSI) determined from a single addition of 150 mg P/100 g soil was related to soil Ca and CaCO₃, but the correlation coefficients were rather low (r = 0.46 and 0.38, respectively). A P availability index (PAI), determined from the increase in extractable soil P after adding 50 mg P/kg to a suspension and allowing it to dry, was correlated quite well with cation exchange capacity and clay content (r = ‐0.61 for each) in soils with pH < 8.8. The PAI also had a positive relationship with the density of the processed soil sample (r = 0.60). The relationship between PAI and soil Ca (r = ‐0.51) was also better than that between PSI and soil Ca. Inclusion of initial soil P and organic carbon along with CEC increased the predictability of PAI from 37% to 59%. In soils with pH > 8.8, soil pH was the dominant factor controlling the PAI (r = 0.92).     

Soil test calibration studies for formulation of phosphorus fertilizer recommendations for maize in Morogoro district,Tanzania. II. estimation of optimum fertilizer rates

Abstract

Studies were conducted on some soils of Morogoro District to estimate economically optimum phosphorus (P) fertilizer recommendations for maize from soil tests. The studies involved the estimation of maize yield response to added P in pot and field experiments. Maize responded to added P at three sites. At all three sites, the residual effects of added P lasted up to three years suggesting that added P was not strongly fixed in these soils. Yield responses were related to extractable P contents by the Olsen and AB‐DTPA methods through a modification of the Mitscherlich‐Bray equation. Estimated model parameters were incorporated in equations for estimating optimum P fertilizer rates for different cost‐price ratios (p) and marginal rates of return (R). At the p and R values that prevailed in the study area during 1997, recommended P fertilizer rates (P_R) could be determined by the equation: P_R=68.497–1.191T (Olsen P). Calculated P fertilizer rates for P deficient sites in the district ranged from 15 to 60 kg P ha^‐1. This wide range underscored the importance of site specific fertilizer recommendations.     

Phosphate sorption in relation to aluminum and iron oxides of oxisols from Ghana

Abstract

Phosphate (P) sorption characteristics of six natural Ghanaian Oxisols, selected because of their hydrological and topographical suitability for agriculture, were evaluated. Availability of P appears to be adequate for half of the soils as suggested by the Bray P1 test and determination of the standard P requirement (SPR), i.e., the amount of P sorbed at a concentration of 0.2 ppm P (6.46 μM). The SPR was found to be very closely related to P_max (Langmuir P sorption capacity), which in turn, was significantly correlated with oxalate‐extractable aluminum (Al) (Al_o) and iron (Fe) (Fe_o) and related (not significantly) to the difference between dithionite‐citrate‐bicarbonate‐extractable Fe (Fe_d) and oxalate‐extractable Fe. Accordingly, P_max is fairly well predicted by the model of Borggaard: P_ca]e=0.211#lbÀl_o+0.115#lbFe_o+ 0.05#lb(Fe_d‐Fe_o)+0.3, except for one soil strongly enriched in Fe oxides, mainly goethite. This goethite was found by X‐ray diffraction analysis to consist of crystals larger than normally found for pedogenic Fe oxides. The difference between P_max and P_calc for this soil could, therefore, be attributed to the occurrence of these large Fe oxide crystals, because P sorption will decrease with increasing crystal size (decreasing specific surface area).     

Assessing extractable soil phosphorus methods in estimating phosphorus concentrations in surface run‐off from Calcic Hapludolls

Understanding soil test phosphorus (STP) and surface run‐off phosphorus (P) relationships for soils is necessary for P management. The objective of the study was to evaluate the efficacy of various soil test indices to predict P losses in surface run‐off. Selected sites were subjected to in situ rainfall simulations according to the protocol of the National Phosphorus Research Project ( NPRP, 2001 ). P from a composite of twenty‐four 2.0‐cm‐diameter core soil samples (0–5 cm) was extracted using the Olsen, Bray–Kurtz, Mehlich III, distilled water and 0.01 m calcium chloride procedures. All of these P extraction methods explained a significant amount of variability in surface run‐off total dissolved P [TP (<0.45)] (r² ≥ 0.67; P ≤ 0.01), where 0.45 is the filter pore diameter in microns. Multiple regression models showed extractable P to be the best soil predictor of surface run‐off TP (<0.45) among the studied soils. Despite extraction method or soil type, extractable P was the best soil predictor of surface run‐off TP (<0.45). Either agronomic (0.92 ≤ r² ≤ 0.96) or environmental (0.94 ≤ r² ≤ 0.96) soil tests were effective in estimating surface run‐off TP (<0.45) in select Mollisols.     

The effect of soil flooding on the transformation of Fe oxides and the adsorption/desorption behavior of phosphate

As repeatedly reported, soil flooding improves the availability of P to rice. This is in contrast with an increased P sorption in paddy soils. The effects of soil flooding on the transformation of Fe oxides and the adsorption/desorption of P of two paddy soils of Zhejiang Province in Southeast‐China were studied in anaerobic incubation experiments (submerging with water in N₂ atmosphere). Soil flooding significantly increased oxalate‐extractable Fe (Fe_ox), mainly at the expense of dithionite‐soluble Fe (Fe_DCB), as well as oxalate‐extractable P (P_ox), but decreased the ratio of P_ox/Fe_ox. Flooding largely increased both, P adsorption and the maximum P adsorption capacity. The majority of newly sorbed P in the soils was P_ox, but also more newly retained P was found to be not extractable by oxalate. Flooding also changed the characteristics of P desorption in the soils. Due to a decrease of the saturation index of the P sorption capacity, P adsorbed by flooded soils was much less desorbable than that from non‐flooded soils. There are obviously significant differences in the nature of both, the Fe_ox and P_ox fractions under non‐flooded and flooded conditions. The degree of the changes in Fe_ox, P_ox, P adsorption and P desorption by flooding depended on the contents of amorphous and total Fe oxides in non‐flooded soils. Our results confirm that the adsorption and desorption behavior of P in paddy soils is largely controlled by the transformation of the Fe oxides. The reasons of the often‐reported improved P availability to rice induced by flooding, in spite of the unfavorable effect on P desorbability, are discussed.