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


     

Published online 27 October 2005
Published in Soil Sci Soc Am J 69:2000-2011 (2005)
DOI: 10.2136/sssaj2003.0320
© 2005 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 Similar articles in Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Yampracha, S.
Right arrow Articles by Yost, R. S.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Yampracha, S.
Right arrow Articles by Yost, R. S.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Yampracha, S.
Right arrow Articles by Yost, R. S.
Related Collections
Right arrow Phosphorus
Right arrow Nutrient Management
Right arrow Wetland Soils

Soil Fertility & Plant Nutrition

Predicting the Dissolution of Rock Phosphates in Flooded Acid Sulfate Soils

Sukunya Yamprachaa, Tasnee Attanandanaa, Aminata Sidibe-Diarrab and Russell S. Yostb,*

a Dep. of Soil Science, Faculty of Agriculture, Kasetsart University, 50 Phaholyothin Rd, Jatujak, Bangkok, Thailand
b Dep. of Tropical Plant and Soil Sci., Univ. of Hawaii at Manoa, 3190 Maile Way, Honolulu, HI 96822

* Corresponding author (rsyost{at}hawaii.edu)


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Seven hundred sixty thousand hectares of acid sulfate soils in Central Plain of Thailand are used for rice (Oryza sativa L.) cultivation. Insufficient nutrient P seriously limits rice growth and yield. A local rock phosphate (RP) Kanchanaburi RP (KRP), a reference RP (Gafsa from Tunisia), and a KRP containing soluble P (KRPS) were used to investigate the dissolution and availability in six acid sulfate soils under flooded conditions. The soil properties that have a major influence on RP dissolution and P sorption were investigated for developing a model and algorithms for estimating the RP requirement for rice. High KCl-extractable aluminum (Al_KCl) and low soil pH enhanced the dissolution of KRP, Gafsa, and KRPS. The substantial calcium carbonate equivalent of the RP increased the pH of soils and limited RP dissolution. The P sorption of the soils was estimated using Bray 1 and 2 extractions. The P sorption was apparently greater than RP dissolution a few days after submergence in some acid sulfate soils, decreasing the Bray 1 level. Phosphorus extractable by Bray 2 increased with incubation time where KRP, Gafsa, and KRPS were applied. Amounts of RP predicted using an algorithm based on predicted dissolution and sorption in the various soils were similar to a local estimate of RP requirement.

Abbreviations: Al_KCl, aluminum extracted by KCl • Bray_P2, amount of P extracted by Bray 2 • Cc, Chachangsao • KRP, Kanchanaburi rock phosphate • KRPS, Kanchanaburi rock phosphate containing soluble phosphate • Ma, Maha-pot • NaOH-P, P extracted by NaOH • Ok, Ongkharak • PBray 2, P extracted by Bray-2 extractant • Pc, critical level of soil P for a specific extractant • Rs, Rangsit • Rsa, Rangsit very acid soil • Se, Sena • {Delta}P_Bray 2, change in Bray-2 extractable P • {Delta}P_NaOH, change in NaOH-extractable P


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
ACID SULFATE SOILS occur mainly in the Central Plain of Thailand, most of which are suitable for rice cultivation (Department of Agriculture, 2004). Most of the acid sulfate soils in Central Plain are active acid sulfate soils with soil pH below 4 within the 50-cm depth and contains yellow mottle of jarosite. The acidity is directly or indirectly caused by sulfuric acid formed by the oxidation of pyrite (van Breemen, 1982). Acid sulfate soils in the Central Plain of Thailand are classified into three categories of paddy rice suitability classes (Kevie and Yenmanas, 1972). Soil acidity is one of the criteria used for this classification and depth of jarosite mottles is a visual acidity index. The three categories are: (1) PIIa soils class, with an estimated area of 0.32 million ha and jarosite mottles were found at >100-cm depth. (2) PIIIa soils class, with an approximate area of 0.38 million ha and jarosite mottles were found within the 50- to 100-cm depth. (3) PIVa soils class, with an estimated area of 0.06 million ha and jarosite mottles were found <50-cm depth.

Rice yields are often low as a result of the low potential productivity in acid sulfate soils (Land Development Department, 2002). The low potential productivity is often due to low P availability that limits growth and rice yield (Attanandana and Vacharotayan, 1984). Soluble phosphate fertilizer such as mono-ammonium phosphate (16–20–0) and RP have been recommended for rice grown on P deficient acid sulfate soils at the rate of 16.3 kg P ha–1 (Department of Agriculture, personal communication, 2002). However, the recommendations currently do not consider categories of acid sulfate soils.

Rock phosphate can be applied directly to rice and the residual effect from the first application can provide adequate available P for the next crop of rice (Sassanarakkit, 1982). Rock phosphates are found in the West, North, Northeast, and Southern part of Thailand. Kanchanaburi RP, derived from bat guano from West part of Thailand, is located close to the Central Plain where acid sulfate soils are found (Warinthomnuwat, 1999). Because of low content of total P in some Kanchanaburi RP, the fertilizer manufacturers often add soluble phosphate fertilizer to increase total P.

We suggest that predicting RP requirements should first consider factors affecting dissolution and then those that affect sorption and availability of the dissolved P. The first reaction when RP is added to soils is the dissolution, which depends on chemical and physical properties of the RP and soil characteristics such as soil acidity, clay content, organic matter, Al and Fe contents, and Ca and P status (Chu et al., 1962; Hammond et al., 1986; Kanabo and Gilkes, 1987; Bolan and Hedley, 1990). High soil Ca status usually reduces RP dissolution and is frequently measured as Ca saturation, exchangeable Ca, and Ca exchangeable capacity (Babare et al., 1997; Wright et al., 1992; Hammond et al., 1986; MacKay et al., 1986).

Other soil properties that control the dissolution of RP include soil pH. Chu et al. (1962) suggested that RP dissolved more in soils with low pH and was related to free Fe oxide content. Chien et al. (1980) found that the dissolution of North Carolina RP was related to reactive Al while reactive Al was also significantly related to P sorption capacity. In flooded acid sulfate soils, soil pH during submergence is around 4.5 to 5 (K. Kyuma, personal communication, 2001), which is sufficiently acid to dissolve most RP. The soils having the highest clay and free Fe2O3 contents have the highest P sorption capacity (Smyth and Sanchez, 1982). High P sorption capacity enhances the dissolution of RP by reducing the P concentration in soil solution (Smyth and Sanchez, 1982; Syers and MacKay, 1986). In flooded acid sulfate soils, increases in active Fe and Al led to increased P sorption (Jugsujinda et al., 1995).

When the RP is applied to soils and dissolved, an important reaction is the sorption of the dissolved P in the soil (Sample et al., 1980). Generally the P availability increases in flooded soil due to the reduction of insoluble ferric P to more soluble ferrous phosphate and release of occluded phosphate by reduction of hydrated ferric oxide coatings and hydrolysis of Al and Fe phosphate (K. Kyuma, personal communication, 2001; Ponnamperuma, 1972). Moreover, high levels of total C, extractable P, P in soil solution and soil pH tend to reduce P sorption under reduced conditions (Khalid et al., 1979; Sah et al., 1989). Shahandeh et al. (1994) observed that soils with higher oxalate-extractable Fe sorbed more P under reduced conditions. In some cases, however, P sorption in acid sulfate soils under reduced conditions was higher than under oxidized conditions due to increased soil pH (Jugsujinda et al., 1995). The removal of exchangeable, amorphous, and crystalline forms of Fe, Al, and Mn reduced the P-sorption in an acid sulfate soil of Thailand (Jugsujinda et al., 1995). Patrick and Khalid (1974) and Khalid et al. (1977) suggested that anaerobic soils release P into the soil solution when solution P is low and sorb P from soil solution when soils have high solution P. The greater surface area of the reduced ferrous compounds in anaerobic soils results in more soil P being solubilized where solution P is low. However, more solution phosphate is sorbed where solution phosphate was high (Patrick and Khalid, 1974). Krairapanond et al. (1993) also found that the P sorption capacity in acid sulfate soils of Thailand increased with increasing soil pH and decreasing EH. Iron oxide was the primary factor increasing P sorption and Al oxide seemed to play a secondary role on flooded acid sulfate soils (Jugsujinda et al., 1995).

Many P decision support systems have been developed to assist the P nutrient management. The Phosphorus Decision Support System (PDSS) was designed to assist in the diagnosis and correction of P deficiencies in soils and crops, with emphasis on tropical conditions (Yost et al., 1992). The PDSS uses the buffer coefficient, the soil critical level of extractable P, and the current extractable P level to estimate crop P requirement (Yost et al., 1992). However, PDSS was developed for water-soluble P fertilizer. With adaptation of PDSS structure it might be used to estimate crop RP requirements.

A hypothesis of this study is that the dual processes of dissolution and sorption determine the plant availability of RP and provide a useful structure for an algorithm to estimate the amount of RP that should be added to the soil. The objectives of this study were (i) to determine which properties of acid sulfate soils have a major influence on RP dissolution; (ii) to determine which factors affect P sorption; and (iii) to develop a tentative model which apply from PDSS and algorithm that represents these two processes to estimate the RP requirements for rice cultivation in acid sulfate soils of Thailand. Later field-testing, adjustment, and validation are sure to be needed.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Analysis of Soils and Rock Phosphate
Soils
Six acid sulfate soils, representing the most important acid sulfate soils in Thailand, were selected for these experiments: Rangsit very acid (Rsa), Ongkharak (Ok), Rangsit (Rs), Sena (Se), Maha-phot (Ma), and Chachangsao (Cc) soil series, represent soils from the Central Plain of Thailand. The Maha-phot, and Chachangsao soil series belong to PIIa soils class, Se and Rs soil series belong to PIIIa soils class. The soils in PIVa class are the Ok and Rsa soil series (Land Development Department, 2002). The Se soil collected for this experiment was, however, more acidic than the typical Se soil series (T. Attanandana, unpublished data, 2003). Surface (0–10 cm) soil was collected from each location, air-dried, sieved (<2 mm), and thoroughly mixed. Chemical characteristics of the soils are summarized in Table 1. Soil pH was determined by a glass electrode (soil/water ratio of 1:1), exchangeable cations were determined in an extraction with ammonium acetate (pH 7). Extractable Al was analyzed by three methods, (i) ammonium oxalate, pH 3 (Jackson et al., 1986); (ii) 1 M KCl extractions (Barnhisel and Bertsch, 1982); and (iii) dithionite extraction (Asami and Kumada, 1959). Extractable Fe was measured by two methods: ammonium oxalate, pH 3 (Jackson et al., 1986), and dithionite (Asami and Kumada, 1959). Phosphorus was extracted by Bray 1 and 2 (Bray and Kurtz, 1945), ammonium oxalate, pH 3 (Guo and Yost, 1999), and as P sorption capacity (Saunders, 1965). Effective cation exchange capacity was the summation of exchangeable bases and KCl-extractable Al (USDA, 1995).


View this table:
[in this window]
[in a new window]
  		Table 1. Chemical properties of acid sulfate soils from central Thailand.

 
Rock Phosphates
Kanchanaburi RP, Gafsa RP, and KRPS were compared. The KRPS is a mixture of Kanchanaburi RP and mono-ammonium phosphate fertilizer (MAP) (16–20–0) at the rate of 2:1 by weight. Kanchanaburi RP was mixed with mono-ammonium phosphate fertilizer by fertilizer manufacturers to increase total P.

Total P, Ca, and water-soluble P of the RP were measured by the method of AOAC (1984) and Official Methods of Analysis of Fertilizer (The National Institute of Agricultural Sciences Ministry of Agriculture, Forestry, and Fisheries, Japan, 1982) (Table 2). Phosphorus solubility of Kanchanaburi, Gafsa, and KRPS was represented by a second 2% (w/v) citric acid, a first 2% (v/v) formic, and a second neutral ammonium citrate extraction. Citric, formic, and neutral ammonium citrate P solubility of Kanchanaburi and Gafsa were also determined by International Fertilizer Development Center (IFDC) (U. Singh, personal communication, 2003). Calcium carbonate equivalent (CCE) was measured by digesting RPs with 0.50 M HCl and back titrated with 0.25 M NaOH to pH 5 (Sikora, 2002). Titration to pH 5.0 was an adaptation suggested by Sikora (2002), to improve the accuracy of estimating CCE of RP. The particle-size distribution analysis was performed using sieves of 0.05, 0.15, and 1 mm.


View this table:
[in this window]
[in a new window]
  		Table 2. Chemical characteristics of Kanchanaburi and Gafsa rock phosphate.

 
Predicting Rock Phosphate Dissolution
The KRP, KRPS, and Gafsa RP materials were added to the six acid sulfate soils at the rates of 0 and 500 mg P kg–1 with three replications for each sampling time. One gram of soil mixture was weighed into a screw cap tube. The soil mixture was flooded with 30 mL of distilled water and incubated at 30°C for 56 d. The soil mixtures in a screw cap tubes were sampled after 0, 3, 5, 7, 14, 28, 42, and 56 d of incubation. Ten milliliters of 4 M NaCl were added to flooded soils (NaCl concentration was 1 M NaCl and soil/solution ratio was 1:40) and shaken 1 h for prewashing the soil mixture. After prewashing, the soil was shaken with 100 mL of 0.5 M NaOH for 16 h to estimate the dissolution of RPs (MacKay et al., 1986). The P content in the extracted solutions was determined using the method of Murphy and Riley (1962).

The differences in P levels ({Delta}P) from the dilute NaOH extraction between RP-treated and RP-untreated samples were used as an estimate of RP dissolution while sorption was estimated with Bray-1 and Bray-2 extractions

Predicting Rock Phosphate Sorption
Screw cap tubes containing 10 g of the soil mixtures for each sampling time were flooded with 25 mL of distilled water. The soil mixtures were incubated and sampled under the same conditions and times as in the dissolution study. The soils were extracted with Bray 1 and 2 solutions to estimate the sorption of dissolved RPs (Bray and Kurtz, 1945). The concentration of Bray 1 is 0.025 M HCl and 0.03 M NH4F but the soil in this study flooded with water so solutions were prepared such that the final Bray-1 concentration matched that of the Bray-1 method. Bray-2 extractable P was prepared in the same way as the Bray-1 extraction so that the final concentration was that specified by the Bray-2 procedure. The soil mixtures were extracted with Bray 2 at soil and solution ratio 1:10 for 1 min. The P content in the extracted solutions was determined using the method of Murphy and Riley (1962).

The differences in P level from the Bray 1 or 2 from soil mixtures with and without RP were used as an estimate of RP sorption. The pH (soil/water ratio 1:2.5) was measured during the incubation period. Bray-2 extraction is used widely in Thailand because it is significantly correlated with crop response. However, strong acidity in both extractions can dissolve an amount of undissolved RP. Then the RP availability was overestimated.

Phosphorus Buffering Coefficients in Flooded Acid Sulfate Soils
Ten grams of each of the acid sulfate soils were amended with potassium hydrogen phosphate (KH2PO4) at the rate 0, 25, 50, 100, 200, and 400 mg P kg–1. The soil mixtures were flooded with 25 mL of distilled water and incubated at 30°C for 2 wk. The soils were extracted with Bray 1 and 2 in the same way of RP sorption study. Phosphorus buffer coefficients (PBC) were estimated from the slope of the linear regression of extractable soil P regressed against added P.

Statistical Analysis and Model Development
Developing a Model for Predicting Rock Phosphate Dissolution
Changes in {Delta}NaOH-P over time were analyzed for significant differences from previous points in time by using repeated measures in SAS. The differences of least square means were computed by using Tukey-Kramer adjustment method.

Changes in NaOH-P over time were modeled using a Mitscherlich equation as follows:

[1]
Where Y({Delta}P_NaOH) = (amount of NaOH-P in the soils + rock phosphate) – NaOH-P in soil alone, mg P kg–1

The changes in {Delta}P_NaOH in KRP, Gafsa RP, and KRPS over time in acid sulfate soils were used to fit the Mitscherlich equation (Eq. [1]) for each soil by using the Statistical Analysis System program (PROC NLIN) (SAS, 1985). The residual mean square (RMS) was computed to indicate the goodness of fit. The estimated Coefficient A from Mitscherlich fit indicates the maximum RP dissolution and was correlated with properties of six acid sulfate soils to explain the relationship between RP dissolution and soil properties.

The Coefficients A, B, and C from Mitscherlich fit of five acid sulfate soils were then regressed on soil properties to determine which soil properties affected the coefficients, using a stepwise regression procedure in SAS. The regression between A, B, and C coefficient of five acid sulfate soils and soil properties of five acid sulfate soils were used to estimate the A, B, and C coefficient in other soil. Finally, the dissolution of KRP, Gafsa, and KRPS in soil was estimated by substituted A, B, and C coefficients into Eq. [1] (model fit) for each of the six soils. The RMS was computed to indicate the goodness of fit.

Developing an Algorithm to Estimate the Amount of Rock Phosphate to be Added
The P prediction model implemented in PDSS uses the following equation to predict P requirement (Chen et al., 1997):

[2]
where Preq = the fertilizer P requirement








Predicting Phosphorus Buffering Coefficient
Stepwise regression procedure in SAS was used to regress the P buffering coefficient on soil properties to determine which soil properties affected the P buffering coefficient.


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Rock Phosphate Dissolution
Kanchanaburi Rock Phosphate (KRP)
Small amounts of P (0.006%) in KRP were extracted by 0.5 M NaOH. The increase in NaOH-extractable P (P_NaOH) in the KRP-treated soils relative to the KRP-untreated soils was used to estimate the amount of P dissolved from the KRP ({Delta}P_NaOH). The KRP dissolution in six acid sulfate soils after 0 d was approximately 100 to 150 mg P kg–1 with more dissolved in Rsa, Ok, and Se soils (Fig. 1) . The rate of dissolution after 56 d of incubation increased to 47 to 54% of the total P added. The KRP dissolved only to a small extent in Rs, Ma, and Cc soils. It appears that increasing soil pH strongly limited KRP dissolution. The soil pH in Rsa, Ok, and Se soils amended with KRP (0 d after incubation) increased only about 0.5 pH units to 4.5, after phosphate addition. The Rs, Ma, and Cc soils, in contrast, when amended with KRP (0 d after incubation), increased to pH 5.5. This substantial increase in soil pH seems to be a result of the high CCE of KRP (Table 2) and the reduction of soil after flooding. The KRP dissolution in six acid sulfate soils, therefore, appears to have been limited by reduced soil acidity in soil system after adding KRP. Attanandana and Vacharothayan (1984) observed that the simultaneous application of RP and lime in paddy soils increased soil pH to 5.5 and reduced the availability of RP. The KRP dissolution in Rsa, Rs, and Se soils was not significant difference after 14 d of incubation (Table 3), which might suggest that KRP dissolution in Rsa, Rs, and Se soil reached a plateau after 14 d of incubation. Kanchanaburi RP dissolution in Ok soil was not a significant difference after 28 d of incubation; it might be suggested that KRP dissolution reached a plateau after 28 d of incubation. The KRP dissolution in Ma and Cc soils shows a significant difference until the end of incubation.



View larger version (42K):
[in this window]
[in a new window]
  		Fig. 1. Dissolution of Kanchanaburi rock phosphate (KRP) as represented by the differences in NaOH-extractable-P, Mitscherlich fit curve and changes of soil pH between KRP-amended (KRP 500 mg P kg–1) and KRP unamended soils (KRP 0 mg P kg–1) in acid sulfate soils initially and after amending with KRP.

 

View this table:
[in this window]
[in a new window]
  		Table 3. The difference of RP dissolution between time of incubation.

 
The change in {Delta}P_NaOH over time where KRP was applied to six acid sulfate soils was fitted with a Mitscherlich equation (Eq. [1]) to describe the dissolution. The resulting prediction from the Mitscherlich equation for each soil was plotted with the experimental data (Fig. 1). The RMS of KRP dissolution in six acid sulfate soils as fitted by Mitscherlich equation was below 16.1 except KRP dissolution in Rsa soil (Table 4).


View this table:
[in this window]
[in a new window]
  		Table 4. The equation for described the RP dissolution in six acid sulfate soils as fitted by Mitscherlich and model which was developed from soil properties and residual mean square.

 
Gafsa Rock Phosphate
The initial change in NaOH-P where Gafsa RP was applied to six acid sulfate soils was less than with KRP or approximately 50 to 100 mg P kg–1 (Fig. 2) . The dissolution at initial time may have been less than for KRP dissolution because 64% of the KRP has a particle size <0.15 mm while only 19% of the Gafsa RP was smaller than 0.15 mm (Table 2). The KRP and Gafsa RP were not ground before incubating; the dissolution of highly reactive unground RP may be different initially but after reaching equilibrium they were not different (Chien, 1998).



View larger version (43K):
[in this window]
[in a new window]
  		Fig. 2. Dissolution of Gafsa rock phosphate (RP) as represented by the differences in NaOH-extractable-P, Mitscherlich fit curve and changes of soil pH between Gafsa RP-amended (Gafsa RP 500 mg P kg–1) and Gafsa RP unamended soils (Gafsa RP 0 mg P kg–1) in acid sulfate soils.

 
More of the Gafsa RP dissolved during the flooding than KRP probably due to the pH of six acid sulfate soils remained lower than the pH of soils amended with KRP (Fig. 2). The small increase in pH of six acid sulfate soils after Gafsa RP addition (0 d after incubation) was less than the increase in soil pH after flooding. Increasing pH was an expected effect of reduction and likely a limiting factor to the dissolution of Gafsa RP for some soils. Gafsa RP dissolved in Rs, Ma, and Cc soils was not significant difference after 42 d (Table 3), then the Gafsa RP dissolution in Rs, Ma, and Cc soils reached a plateau at 42 d. The Gafsa RP dissolution in the more acid Rsa, Ok, and Se soils was not a significant difference after 28 d of incubation (Table 3) while the soil pH remained at pH 5.0 or lower, even with flooding. The high solubility of Gafsa RP may be a factor that promoted the greater dissolution of Gafsa RP than KRP. A second neutral ammonium citrate extraction in Gafsa RP also indicated Gafsa RP had higher solubility than KRP (Table 2).

The changes in {Delta}P_NaOH of Gafsa RP over time in six acid sulfate soils were also fitted with a Mitscherlich equation and plotted with experimental data (Fig. 2). The RMS of Gafsa RP dissolution in Rsa, Ok, and Rs soils as fitted by Mitscherlich equation was higher than other soils (Table 4).

Kanchanaburi Rock Phosphate with Soluble Phosphorus
The KRPS contained around 1.4% soluble P when extracted by NaOH or 0.71% P when extracted with water (Table 2). This means that about 70 mg of soluble P kg–1, as extracted by NaOH, should have been detected where KRPS was applied at the rate 500 mg P kg–1. It may reflect the possibility that the MAP mixed with KRP may have contained Al-Fe-P impurities. Since, the KRPS contained KRP and MAP in a ratio of 2:1 by weight, therefore KRPS should contain around 140 mg P kg–1 soluble P when KRPS was applied at a rate of 500 mg P kg–1. It might be suggested that the loss of 70 mg P kg–1 of MAP probably precipitated as Ca-P from MAP and free CaCO3 of KRP during mixing KRP with MAP.

The dissolution of KRPS in six acid sulfate soils at time 0 of incubation was high (Fig. 3) . The pH of six acid sulfate soils after the addition of KRPS increased only slightly in some soils. When KRPS dissolved in Ok soil, {Delta}P_NaOH was nearly constant after 28 d of incubation because there was no significant amount of KRPS in Ok soil after 28 d of incubation (Table 3). The KRPS dissolution in Rsa, Se, Rs, and Cc soils continuously increased until the end of incubation while KRPS dissolution in Ma soil was not a significant difference after 14 d of incubation (Table 4). The KRPS dissolution thus could be described relatively well by a Mitscherlich function as in the case for KRP and Gafsa RP dissolution (Table 4).



View larger version (42K):
[in this window]
[in a new window]
  		Fig. 3. Dissolution of Kanchanaburi rock phosphate with soluble phosphate (KRPS) as represented by the differences in NaOH-extractable-P, Mitscherlich fit curve, and changes of soil pH between KRPS-amended (KRPS 500 mg P kg–1) and KRPS unamended soils (KRPS 0 mg P kg–1) in acid sulfate soils.

 
Predicting Rock Phosphate Dissolution
The A coefficient of Mitscherlich fit, which indicates the maximum dissolution of KRP in six acid sulfate soils, was highly correlated with soil properties (Table 5). The soil acidity before incubation as represented by soil pH, KCl-extractable Al, and Al saturation percentage, was significantly related to the maximum dissolution (Coefficient A) of KRP, Gafsa, and KRPS. The maximum dissolved KRP, Gafsa, and KRPS increased with increasing soil acidity as measured by extractable Al, Al saturation, KCl-extractable Al and decreasing soil pH has also been suggested by many others (Chu et al., 1962; Bolan and Hedley, 1990; Wright et al., 1992; Kanabo and Gilkes, 1987). Soil acidity increased RP dissolution in all six acid sulfate soils. The extractable Al increased with decreasing soil pH. The dissolution of RP appears to increase when the soil pH decreases because of the protons that become available for dissolution of RPs (Bolan and Hedley, 1990).


View this table:
[in this window]
[in a new window]
  		Table 5. The correlation between maximum dissolution of KRP, Gafsa and KRPS (A coefficient) and properties of six acid sulfate soils.

 
The soil properties that were highly correlated with maximum dissolution of KRP, Gafsa, and KRPS were entered in a multiple regression to attempt to predict the A, B, and C coefficients of each acid sulfate soil.

Predicting Dissolution of Kanchanaburi Rock Phosphate
The regression between A, B, and C coefficient and soil properties indicated the dissolution of KRP increased with increasing extractable aluminum. The B value, the difference between the asymptote and intercept, increased with increased extractable Al. A high B value represents less dissolution of KRP at time 0 of the incubation. Decreasing extractable Al led to an increased C value. The example of regression between A, B, and C coefficient and soil properties for predicting A, B, and C coefficient of KRP dissolution in Rs soil were giving the following equation:

[3]

[4]

[5]

The coefficients for testing the KRP dissolution in each soil substituted into Eq. [1] to predict KRP dissolution. The prediction of KRP dissolution in six acid sulfate soils from model fit as based on soil properties were calculated the RMS for comparison the goodness of fit with Mitscherlich equation (Table 4). The RMS of model fit as prediction KRP dissolution in Rsa and Ok soils was higher than RMS of Mitscherlich fit while another soils was not much difference.

Predicting Dissolution of Gafsa Rock Phosphate
The regression between A, B, and C coefficient and soil properties indicated the soil acidity in acid sulfate soils was important effect on Gafsa RP dissolution. The Gafsa RP dissolution increased with decreased soil pH. Decreasing extractable Al led to an increased C value. The amount of soil acidity in acid sulfate soils was important in predicting both Gafsa RP dissolution and KRP dissolution. The example of regression between A, B, and C coefficient and soil properties for predicting A, B, and C coefficients of Gafsa RP dissolution in Rs soil were given in the following equation:

[6]

[7]

[8]

The predicted values of A, B, and C coefficient were also substituted into Eq. [1] to predict Gafsa RP dissolution in six acid sulfate soils. The prediction of Gafsa RP dissolution in Ok, Rs, Ma, and Cc soil was not close to the experimental data based on the high RMS in those soils (Table 4). RMS of predicting Gafsa RP dissolution in Rsa and Se soils by model fit was no different than with RMSE of Mitscherlich fit.

Predicting Dissolution of Kanchanaburi Rock Phosphate Containing Soluble Phosphate
Dissolution of KRPS was also estimated by fitting Eq. [1] to the dissolution and then determining which soil properties predicted the fitted A, B, and C values. Stepwise multiple linear regressions between the fitted A, B, C coefficients and soil properties were also obtained from the incubation of five acid sulfate soils with KRP containing soluble P for 56 d. The dissolution of KRPS increased with increasing Al_KCl. The B value increased with increased Al_KCl of soils. Increasing soil pH led to an increased C value. A high C value indicates that the dissolution of KRPS reached equilibrium rapidly. The example of regression between the A, B, and C coefficients and soil properties for predicting the A, B, and C coefficients of Gafsa RP dissolution in Rs soil were giving the following equation:

[9]

[10]

[11]

The prediction of KRPS dissolution in Ok soil by model fit revealed RMS close to the Mitscherlich fit while the RMS of the prediction of KRPS dissolution in another soils was higher than for the Mitscherlich fit (Table 4).

Phosphorus Sorption
Kanchanaburi Rock Phosphate
The P extractability from KRP, Gafsa RP, and KRPS was estimated by both Bray 1 and Bray 2 (Fig. 4 and 5) . The P extractability from KRP in Rsa, Ok, and Se soils with Bray 1 at 0 to 7 d increased and became relatively stable after 14 d of incubation (Fig. 4). The P extractability in Rs, Ma, and Cc soils, however, continued to slightly increase. The P extractability by Bray 2 in six acid sulfate soils increased from 0 to 7 d and decreased little after 14 d of incubation (Fig. 5).



View larger version (16K):
[in this window]
[in a new window]
  		Fig. 4. Phosphorus extractability from rock phosphate in acid sulfate soils as presented by the differences in Bray-1 extractable P content between RP-amended (RP 500 mg P kg–1) and RP unamended soils (RP 0 mg P kg–1)

 


View larger version (18K):
[in this window]
[in a new window]
  		Fig. 5. Phosphorus extractability from RP in acid sulfate soils as presented by the differences in Bray-2 extractable P content between RP-treated (RP 500 mg P kg–1) and RP untreated soils (RP 0 mg P kg–1)

 
The decreasing Bray-1 P extractability after a few days of incubation may be due to several factors including re-adsorption on newly formed Fe2+ and residual hydrous oxide phases as well as by P precipitation of Fe2+ and Ca2+ (Holford and Patrick, 1979). The much higher acid concentration of the Bray-2 extractant seems to reduce subsequent decreases. The P sorption capacity increased in reduced soil at the final soil pH around 6.4 (Holford and Patrick, 1979) and with a high level of added P (Khalid et al., 1977). The pH of Rs, Ma, and Cc soils increased to 6.5 to 7 and the extractability by Bray 1 in those soils was reduced.

Gafsa Rock Phosphate
The P extractable by Bray 1 where Gafsa RP had been applied to six acid sulfate soils was similar to KRP extractability except in the Ok and Rs soils (Fig. 4). The Bray 1-extractable P in Ok and Rs soils increased from 0 to 7 d and then decreased after 14 d of incubation. In contrast, the P extractability by Bray 2, with the higher acid concentration, increased with time in most soils except in Ok soils where the extractability decreased after 14 d of incubation.

Kanchanaburi Rock Phosphate with Soluble Phosphate
The extractability of P by Bray 1 where KRPS had been applied in six acid sulfate soils reached the maximum at 3 to 7 d of incubation. After that the P extractability decreased (Fig. 4). Perhaps the decrease in Bray-1 extractable P after 7 d of incubation occurred because of high P content in soil solution from the soluble P in KRPS led to re-adsorption.

The extractability of P from KRPS in Rsa, Ok, Se, and Rs soils by Bray 2 also decreased with increasing time (Fig. 5). In Ma and Cc soil, the P extractability initially increased as the incubation time increased and then became constant after 28 d of incubation. The pH of the Ma and Cc soils increased during submergence to neutral and RP dissolution reached a maximum.

An Algorithm to Estimate the Amount of Rock Phosphate to be Added
When the RP is added to the soil, P can be released from the RP but not all of dissolved phosphate will be available to the plant. Several researchers have tried to correlate the relation between dissolution of RP with crop yield and P uptake by plant but the correlation was poor (Syers and MacKay, 1986; Bolan and Hedley, 1990). Phosphorus sorption of the dissolved P from RP apparently plays an important role in P availability by the plant.

The ratio between {Delta}P_NaOH/{Delta}P_Bray 2, is proposed to represent the effect of P sorption by the soil after RP is added and dissolved in the soil. The ratio between {Delta}P_NaOH/{Delta} P_Bray 2 in KRP, Gafsa RP, and KRPS in six acid sulfate soils was different at the initial time of incubation but reached an approximate steady state at 28 d of incubation (Fig. 6) . A similar ratio between {Delta}P_NaOH and Bray 1 or Olsen P appeared to reach a steady state value for aerobic soils in West Africa (Sidibé et al., 2004). The relationship between dissolution and extractability (Bray_P2) can be used to convert a Bray-2 extractable P requirement to a NaOH-extractable P requirement. Then, the RP requirement can be estimated based on the RP dissolution model. From the relationship between dissolution (NaOH-P) and extractability (Bray_P2) we propose estimating the amount of RP that should be added to supply nutrient P from the following equation:

[12]
Where RP added = the RP should be added (kg P ha–1)



View larger version (16K):
[in this window]
[in a new window]
  		Fig. 6. The {Delta}NaOH_P/{Delta}Bray 2 ratios from rock phosphates in acid sulfate soils during submergence.

 
The stepwise regression analysis between PBC as extracted by Bray 2 and soil properties indicated that decreases in P retention, Fe and Al as extracted by oxalate increased the P buffering coefficient (PBC) according to the following equation.


[13]
Then, the PBC of flooded acid sulfate soils can be estimated by substituting values for the soil properties in Eq. [13].

Several cautions in the use of Eq. [12] are noted:


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
Soil acidity was a key factor in the dissolution of RP in flooded acid sulfate soils as estimated by the difference in NaOH-extractable P in soils treated and untreated with RP. The soil pH changes during the flooded period strongly affected RP dissolution as the RP dissolution reached equilibrium when soils pH increased to pH 6 to 6.5. Furthermore the CCE of the RP was an important property because it increased soil pH and appeared to limit dissolution. The Bray-1 extractable P decreased after a few days of incubation in some acid sulfate soils, which may be due to several factors including re-adsorption on newly formed Fe2+ and residual hydrous oxide phases as well as by P precipitation of Fe2+ surfaces and Ca2+. The ratios between {Delta}P_NaOH/{Delta}P_Bray2 at 28 d of incubation have potential for predicting the amount of RP that should be added for rice cultivation in acid sulfate soils. These predictions, however, need to be tested in field conditions.


   ACKNOWLEDGMENTS
 
The authors gratefully acknowledge the financial support provided by the Thailand Research Fund and the SM-CRSP Project of the University of Hawaii, USA for this study. Thanks are also due to Dr. B. Boonsompopphan for assistance in collection of soil samples from the rice fields. We thank J. Chuenrung and P. Likhananont, the Office of Factor of Agriculture Production Research and Development, the Department of Agriculture for the incubator and Kanchanaburi RP, respectively.

Received for publication December 5, 2003.


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





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 Similar articles in Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Yampracha, S.
Right arrow Articles by Yost, R. S.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Yampracha, S.
Right arrow Articles by Yost, R. S.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Yampracha, S.
Right arrow Articles by Yost, R. S.
Related Collections
Right arrow Phosphorus
Right arrow Nutrient Management
Right arrow Wetland Soils

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