SSSAJ Journal of Natural Resources and Life Sciences Education
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) 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 ISI 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 HighWire
Right arrow Citing Articles via ISI Web of Science (3)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Wang, X.
Right arrow Articles by Linquist, B.A.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Wang, X.
Right arrow Articles by Linquist, B.A.
Agricola
Right arrow Articles by Wang, X.
Right arrow Articles by Linquist, B.A.
Soil Science Society of America Journal 64:240-246 (2000)
© 2000 Soil Science Society of America

DIVISION S-4-SOIL FERTILITY & PLANT NUTRITION

Predicting Soil Phosphorus Buffer Coefficients Using Potential Sorption Site Density and Soil Aggregation

X. Wanga, J.M. Jackmanb, R.S. Yosta and B.A. Linquistc

a Dep of Agronomy and Soil Science, Univ. of Hawaii, 1910 East West Road, Honolulu, HI 96822 USA
b 4F-2, 89 Tung-Hwa st. 106, Taipei, Taiwan
c International Rice Research Institute, P.O. Box 933, Manila 1099, Philippines

wangx{at}hawaii.edu


   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 Materials and methods
 Results and discussion
 REFERENCES
 
The phosphorus (P) buffer coefficient, a ratio of the increase in extractable P to the amount of applied fertilizer P, is a source of considerable uncertainty in determining the amount of fertilizer needed to meet crop P requirements. The use of clay as a predictor of the P buffer coefficient has been suggested for soils of similar mineralogy. However, it has not been satisfactory for soils with a wide range of soil mineralogies but relatively high clay content. The objective of this study was to improve the prediction of buffer coefficients using soil characteristics associated with the process of P sorption, such as mineralogy, surface area, and aggregation. The soil P sorption site density, estimated from detailed clay mineralogy, and reactive mass, the fraction of the total soil mass in the surface aggregates where newly added P can be sorbed, were used to predict the buffer coefficient. The P buffer coefficients of 10 soils with a wide range in P sorption were estimated by Mehlich 3, modified Truog, and 0.5 M NaHCO3 extractants for incubation periods of 32 and 180 d. The inclusion of P sorption site density and reactive mass substantially improved predicting the P buffer coefficients when compared with the P buffer coefficients predicted by only soil clay content. Statistical models showed that the P buffer coefficients were negatively correlated with both log of the P sorption site density and reactive mass. Thus, soils with fewer P sorption sites, lower reactive mass, and larger aggregate size will tend to have higher buffer coefficients, indicating that a greater portion of the added P remains plant available.

Abbreviations: GMD, geometric mean diameter • PS0.2, the amount of P sorbed at an equilibrium solution of 0.2 mg P L-1 • SPsite: soil P sorption site density • MASSGMD, reactive mass estimated from Eq. [2]


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
Materials and methods
 Results and discussion
 REFERENCES
 
HIGHLY WEATHERED SOILS in the tropics often have low available-P because of high P retention by Al and Fe oxides and amorphous materials (Fox and Searle, 1978; Sanchez and Uehara, 1980). The addition of fertilizer P is one method used to raise soil available P to crop critical levels. The ability to predict P fertilizer requirements and develop accurate P recommendations is critical to sustainable agriculture and to protecting our environment from detrimental effects of excess P.

Phosphorus requirements are often determined from field experiments designed to calibrate soil tests (Evans, 1987), which usually means determining how much fertilizer is needed to meet crop requirements and to produce the maximum yields. A modeling approach which uses buffer coefficients and critical levels of the extractable nutrient represents an effort to improve the empirical process of calibration of P soil test with experiments. The hypothesis of this approach to estimating P requirements is that fewer costly field experiments are required if buffer coefficients and critical levels can be successfully predicted from basic soil properties (Yost et al., 1992).

The P buffer coefficient is defined as the increase in extractable P, extracted by a soil test method, resulting from the addition of fertilizer P to the soil (Cox, 1994), and is often used to calculate how much P must be added to meet plant P requirements. The phosphorus decision support system (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). Once the buffer coefficient is estimated, subsequent estimates of P requirement should depend only on current soil tests.

Characterization of the buffer coefficient should be possible by quantifying soil properties that determine P sorption. Olsen and Watanabe (1963) observed that clay content was correlated with P sorption capacity. Cox (1994) used soil clay content to estimate the soil buffer coefficient, which has been incorporated into the PDSS (Yost et al., 1992), for soils with a wide range of textures but relatively similar soil mineralogical types. Over a wide range of soil mineralogies but relatively high clay content, however, we found little correlation between the buffer coefficient and clay content but high correlation with the amorphous material (mainly, Al, Fe, Si oxides) content as extracted by acid ammonium oxalate. Jackman et al. (1997) further showed that the density of potential P sorption sites, which is a function of the surface area and surface hydroxyl density of minerals, provided an improved prediction of P sorption in a wide range of soils, indicating the strong effect of clay minerals on P sorption.

Highly weathered soils with low clay activity often are highly aggregated (Uehara and Gillman, 1981). Phosphorus sorption and desorption processes appear to be influenced by soil aggregate size and stability because newly applied P remains on the surface of aggregates (Gunary et al., 1964; Linquist et al., 1997; Wang, 1997). Although shaking and grinding soil to less than 2 mm in preparation for P extraction may homogenize the soil sample, it is observed that only a small proportion of the clay fraction (<2 µm) was obtained from some tropical soils with high clay contents (e.g., Kapaa and Leilehua series) even after sonicating the samples for 15 min (Wang, 1997). Consequently, the large size and high stability of soil aggregates may influence the accessibility of sorption sites in the aggregates to P and the ability of extractants to remove the P from sorption sites in the aggregates. Linquist et al. (1997) attempted to quantify the effect of soil aggregation on P sorption using the concept of "reactive mass", which was defined as a fraction of soil mass on the outside of the aggregate into which applied P is initially sorbed. They also reported that P sorption and P concentration in solution were related to the reactive mass. We, therefore, hypothesize that soil P buffer coefficients could be better estimated by the potential P sorption site density and reactive mass.


   Materials and methods
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Materials and methods
Results and discussion
 REFERENCES
 
Soils
Ten soils with a wide range in P sorption and of significant agricultural importance to Hawaii were selected (Table 1) . All samples were collected from the surface horizon, generally 0 to 15 cm in depth. The soils, except the Andisols, were air-dried and ground to pass a 2-mm sieve. The Kaiwiki was dried only to allow the soil to crumble and pass a 6.3-mm sieve. The Maile was dried only sufficiently to pass a 2-mm sieve.


View this table:
[in this window]
[in a new window]
  		Table 1 Soils used for testing an improved method of estimating P buffer coefficients.{dagger}

 
Soil Aggregate Size Distribution
Soil aggregate size distribution was determined by a wet-sieving method with four replicates (Gardner, 1956). Six sieves with openings of 0.1, 0.177, 0.25, 0.42, 0.84, and 2 mm in diameter were used in this study. A 25-g sample of each soil was wetted by capillary rise without vacuum for 15 min before sieving.

The wet-sieving was performed in sieving tanks. The nest of sieves was oscillated up and down in water at 30 cycles per min for a 30-min period. At the end of the 30 min of oscillation, the sieves were removed and allowed to drain. Aggregates were first dried in an oven at 70°C, and then transferred to aluminum dishes and further dried at 105°C.

Results were expressed as the percentage of aggregates and primary particles remaining on each sieve in the whole sample. The percentages were calculated by means of the oven-dry weight of the sample. Geometric mean diameter (GMD) (Mazurak, 1950) was calculated for each soil as follows:

(1)
where i is the mean diameter of each size fraction, wi is the proportion of the total sample weight occurring in the corresponding size fraction, and n is the number of size fractions.

Phosphorus Sorption by Whole Soil
The amount of P sorbed at an equilibrium solution concentration of 0.2 mg P L-1 (PS0.2, an estimate of P sorption potential, Jackman et al., 1997) was estimated by equilibrating 2 g of soil with 20 mL of 0.001 M CaCl solution containing selected concentration of CaH2PO4 (Fox and Kamprath, 1970). The soil P sorption site density (SPsite) was estimated from the quantity of hydroxyl functional groups on the surface of the crystallite as determined from the structure and predominant exposed faces by quantitative x-ray analysis (Jackman et al., 1997). A close relationship between PS0.2 and SPsite was illustrated by Jackman et al. (1997).

Phosphorus Sorption by Soil Aggregates
All soils except the Maile were separated into five aggregate size fractions by wet-sieving (Gardner, 1956): 0.1 to 0.177, 0.177 to 0.25, 0.25 to 0.42, 0.42 to 0.84, and 0.84 to 2 mm (2–6.3 mm for the Kaiwiki soil). After separation, the aggregate fractions were air-dried at 25 ± 2°C. The Maile soil was not used in P sorption study because the soil sample had been dried. This irreversible drying may reduce P retention and increase aggregation (Lim, 1979).

Phosphorus sorption was determined by the modified Fox and Kamprath method (Linquist et al., 1997) as follows. One gram of each soil sample was added to a 125-mL Pyrex flask. This weight of soil formed a very thin layer on the bottom of the flask, thus uniformly exposing the outer surfaces of large and small aggregates to the bulk solution. Ten milliliters of 0.001 M CaCl2 solution containing Ca(H2PO4)2·H2O was added to the flasks, such that P concentrations were at least 0.2 mg L-1 after the 6 d of equilibration (determined from previous experiments). Two P treatments were used and each treatment was replicated twice. The soil solution mixture was agitated gently (50 rpm) at room temperature for 6 d. Following incubation, solutions were analyzed for P. Phosphorus that disappeared from the solution was considered to be sorbed by the aggregates.

Phosphorus concentration after 6 d of equilibration was plotted against aggregate diameter. From the plot, the decrease of P concentration with decreased aggregate diameter was examined to determine the aggregate size at which the P concentration no longer decreased. This aggregate size was assumed to represent complete penetration of the aggregate by P and, therefore, was an estimate of the depth of P penetration (Linquist et al., 1997).

Soil Reactive Mass
Reactive mass was used to estimate P initially sorbed into the aggregates (Linquist et al., 1997). Assuming that aggregates were spheres of uniform density and of diameter, D, and that P would penetrate into the aggregate to the depth, r0, according to the above estimate, the P reactive mass can be calculated by the following equation:

(2)

For aggregates less than or equal to r0 in radius the calculated reactive mass becomes 1. As aggregate size increases the calculated reactive mass decreases approaching 0. The aggregate diameter (D) of whole soil was estimated from the geometric mean diameter.

Phosphorus Incubation Study and Buffer Coefficients
For the incubation study, each soil was treated with five P levels as powdered calcium monobasic phosphate [Ca(H2PO4)2·H2O] (Jackman, 1994). Four sets of P applications were required because of differing sorption capacity of the soils. The applications were 0, 25, 50, 100, and 200 mg P kg-1 soil for the Haiku, Molokai, Pulehu, Waialua soils; 0, 50, 100, 200, and 400 mg P kg-1 soil for the Halii, Kapaa, Makapili, Wahiawa soils; 0, 100, 200, 400, and 800 mg P kg-1 soil for the Maile; and 0, 250, 500, 1000, and 2000 mg P kg-1 soil for the Kaiwiki soil. The soils were placed in plastic bags and the calcium phosphate was thoroughly mixed with the soil. The appropriate amount of water was added to bring the soil to the estimated field capacity. The samples were incubated at 25 ± 2°C for 180 d while the bags were kept open to allow the soils to dry. When the samples had dried they were rewetted to estimated field capacity and thus put through successive wetting and drying cycles. The moisture content was monitored throughout the equilibration period. The Kaiwiki and Maile soils were not allowed to air-dry but were maintained at a moisture level similar to that normally occurring in the field, which prevented an irreversible change in physical or chemical properties of the hydric materials (Lim, 1979).

Soils were sampled after 32 and 180 d of equilibration, and were analyzed for extractable P with the following extractants: (i) 0.01 M H2SO4 + 3% (NH4)2SO4, modified Truog (m-Truog) (Ayres and Hagihara, 1952); (ii) 0.5 M NaHCO3 at pH 8.5 (Watanabe and Olsen, 1965); and (iii) a mixed solution of 0.2 M acetic acid, 0.25 M ammonium nitrate, 0.015 M ammonium fluoride, 0.013 M nitric acid, and 0.001 M ethylenediaminetetraacetic acid (ETDA) (Mehlich, 1984). Phosphorus concentration in the supernatant was measured with ammonium molybdate and ascorbic acid according to Murphy and Riley (1962) and Watanabe and Olsen (1965).

The phosphorus buffer coefficient was estimated from the slope of the linear regression of extractable soil P regressed against added P. A typical linear relationship between extractable P and added P is illustrated by results from the Kapaa soil (Fig. 1) .



View larger version (26K):
[in this window]
[in a new window]
  		Fig. 1 Relationship between extractable P by the three soil test methods (Mehlich 3, modified Truog, and 0.5 M NaHCO3) and P applied to the Kapaa soil following incubation periods of 32 and 180 d

 
Soil water content at -1.5 MPa was measured by pressure plates by pre-soaking the soil in water overnight.

Statistical Analysis
Multiple linear regressions were used to predict the buffer coefficients from soil properties such as P sorption site density, aggregate size, and reactive mass. Regression coefficients were estimated with S-PLUS (Statistical Science, 1995), and the AIC statistic was used to determine the best fitting model (Akaike, 1974). This statistic, the likelihood version of the Cp statistic, provides a convenient criterion for determining whether a model is improved by dropping or adding a term. A logarithmic transformation of the P sorption site density was used in regression analysis and in the computation of Pearson correlation coefficients.


   Results and discussion
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Materials and methods
 Results and discussion
REFERENCES
 
Soil Aggregation
Soils varied in size distribution of aggregates (GMD). One was highly aggregated, several ranged from 0.3 to 0.5 mm, and others less than 0.3 mm (Table 2) . The Kaiwiki soil, dominated by amorphous materials, was highly aggregated as indicated by the GMD value of 0.837 mm (Table 2). The Haiku, Halii, Kapaa, and Makapili soils, dominant in Al and Fe crystalline minerals, were in the middle range, as indicated by GMDs from 0.3 to 0.5 mm. The other soils, characterized by low clay content or by kaolinite and 2:1 clay-sized minerals, were in a group with low soil aggregation — GMDs less than 0.3 mm. The GMD value of the Maile soil should be greater than 0.387 mm, probably because the soil sample had been dried and passed through a 2-mm sieve. Both soils dry irreversibly (Uehara and Gillman, 1981).


View this table:
[in this window]
[in a new window]
  		Table 2 Phosphorus sorption and aggregation of the selected soils

 
Reactive Mass
Changes in solution P concentration with decreasing soil aggregate diameter were used to estimate the depth of P penetration into aggregates, which was then used for calculating reactive mass. The reactive mass was proposed to better explain the effects of aggregate size on P sorption (Linquist et al., 1997). The depth of P penetration into aggregates was soil specific (Fig. 2) . Phosphorus concentration in solution after 6 d of equilibration decreased with decreasing aggregate mean diameter until 0.214 mm for the Haiku, Halii, Kapaa, Makapili, Wahiawa, Waialua soils, and down to 0.139 for the Molokai soil, and 0.177 mm for the Kaiwiki soil. For the Pulehu soil, P concentrations in solution in smaller aggregate fractions were greater than in larger aggregate fractions, and reactive mass was assumed to be 1.



View larger version (22K):
[in this window]
[in a new window]
  		Fig. 2 Phosphorus sorption as influenced by aggregate size for the Kapaa, Kaiwiki, and Pulehu soils. These three soils were selected to represent the different soil P sorption and aggregation. Legends indicate initial P levels added to aggregates. Error bars represent two standard deviations. Symbol sizes exceed error bars in some cases

 
There were negative correlations between both PS0.2 and MASSGMD, and SPsite and MASSGMD (Table 3) . These negative relationships are probably due to the increase in P sorption with increased Al and Fe oxide content (Jones, 1981; Jackman et al., 1997) and a positive correlation between soil aggregation and Al and Fe oxides (Uehara and Gillman, 1981). The highly aggregated soils associated with high contents of Al and Fe oxides would result in smaller values of reactive mass based on Eq. [2], suggesting that the above negative correlations are expected. In contrast, P sorption by aggregates separated from the same soil decreased with increased aggregate size, and increased with increased reactive mass (Linquist et al., 1997; Wang, 1997). This can be expected because the dominant factor affecting P retention is probably reactive mass while the other soil properties are relatively uniform for aggregate fractions separated from the same soil. Similar contents of Al and Fe oxides among aggregate fractions should result in small differences in P retention among aggregate fractions, except for the aggregate size effect of increasing reactive mass with smaller aggregates.


View this table:
[in this window]
[in a new window]
  		Table 3 Correlation coefficients among P sorption potential, P sorption site density, aggregate size, and reactive mass

 
Phosphorus Buffer Coefficient
Phosphorus buffer coefficients were estimated from the slope of the linear regression of extractable P against P applied for incubation periods of 32 and 180 d in this study (Fig. 1). The paired t-test method was chosen to compared the buffer coefficients for 32- and 180-d equilibration periods. There was a significant difference in the buffer coefficients between the two periods for the 0.5 M NaHCO3 extraction (Table 4) . No significant differences in the buffer coefficient for Mehlich 3 and Modified Truog extractions were found by paired t-tests. For the Mehlich 3 extraction, the buffer coefficients of the Waialua, Pulehu, and Maile samples at 180 d of incubation were greater than the values after the 32 d of incubation (Fig. 3) . The higher value for the Maile sample may have resulted from organic P mineralization from the high soil organic carbon content of this soil (240 g kg-1); the other soils with the larger buffer coefficients were the Mollisols (Pulehu, Waialua series), which have higher pH values of 7.7 and 6.5, respectively. For the modified Truog extraction, the Molokai and Waialua samples exhibited larger buffer coefficient values (less P sorption) at 180 d of incubation; the Molokai soil, an Oxisol, was probably limed because its pH was 7.6. These high buffer coefficients mask the expected trend that buffer coefficients should decrease with incubation time reflecting effects of increased P sorption due to slow reactions with the soil. These results again show that use of modified Truog and Mehlich 3 extractants is inappropriate for soils with high pH values because carbonates and alkaline materials neutralize the acid and reduce the effectiveness of the extractants.


View this table:
[in this window]
[in a new window]
  		Table 4 Comparison of P buffer coefficients determined after the 32nd and 180th day of incubation using a paired t-test

 


View larger version (38K):
[in this window]
[in a new window]
  		Fig. 3 Change in P buffer coefficients with time for the Mehlich 3 and modified Truog extractants

 
Decrease in Buffer Coefficients with Time
Several mechanisms have been proposed to explain the slow reaction of P in soils, such as a bonding change with time (Kafkafi et al., 1967) and solid-state diffusion (Barrow, 1983). Recent studies have attributed the slow reaction to P diffusion into sorption sites in micropores of Fe and Al hydrous oxides (Torrent et al., 1990, 1992). Micro- and mesoporosity was found to enhance the slow reaction by lepidocrocite (Cabrera et al., 1981; Madrid and de Arambarri, 1985) and is likely to have the same effect on natural goethites (Parfitt, 1989). Willett et al. (1988) studied P sorption in ferrihydrite and attributed the slow reaction to the migration of P into surface sorption sites of decreasing accessibility within the ferrihydrite particles. These results suggest that for soils incubated with applied P, the initial buffer coefficients may reflect the sorption that takes place at aggregate surfaces, and subsequent decrease in the buffer coefficient with time may reflect the diffusion of P into the aggregates. Shaking soil suspensions destroys aggregates, exposes surfaces, and promotes rapid sorption. Thus we predict that soils with larger and more stable aggregates should have higher buffer coefficients for longer periods of time than the same soils but with smaller aggregates. This may explain why the Maile and Kaiwiki soils, which are rich in amorphous materials and have strong structure, show higher buffer coefficients than some of the Oxisols, which have much lower P sorption potentials estimated by the Fox and Kamprath (1970) P sorption methods. The Fox and Kamprath (1970) method of vigorous shaking usually breaks up aggregates and probably represents maximum P sorption (Linquist et al., 1997). In our case, the Maile and Kaiwiki soils sorbed the most of P according to the vigorous, aggregate crushing equilibration (Table 6) .


View this table:
[in this window]
[in a new window]
  		Table 6 Comparison of means of the P buffer coefficients to P sorption

 
Prediction of P Buffer Coefficients from Soil Properties
Preliminary analysis showed that the buffer coefficients in this study were not correlated with clay content, CEC, oxalate-extractable amorphous materials, pH, Al, organic content, specific surface area, water content at matrix potential of -1.5 MPa, and total P (Jackman, 1994). In this paper, we related the P buffer coefficients to P sorption and other soil properties associated with the process of P sorption. The buffer coefficients were correlated with P sorption potential, aggregate size, and reactive mass, and significantly correlated with log of SPsite (Table 5) . These correlations suggested the presence of compound effects of soil minerals and chemical and physical factors on P sorption and extraction processes in the incubation study. The types of amounts of soil minerals determine the number of sorption sites, which are closely related to the total P sorption potential (Jackman et al., 1997). On the other hand, soil physical properties influence P diffusion into aggregates through moisture retention, aggregate size, and stability. The reactive mass may represent such physical effects on P sorption. Thus both chemical and physical aspects of sorption seem important in estimating buffer coefficients.


View this table:
[in this window]
[in a new window]
  		Table 5 Correlation coefficients describing relationships of the buffer coefficient from three methods (Mehlich 3, m-Truog, 0.5 M NaHCO3) and incubation periods of 32 and 180 d to P sorption potential, log of the P sorption site density, aggregate size, and reactive mass

 
A comparison of the buffer coefficients with PS0.2 values indicates that the measurements of P sorption by the incubation and the P isotherm curve methods were not the same (Table 6). In some cases where there were large differences in PS0.2 values between soils, there was only small or no differences among the buffer coefficients. The Kaiwiki, Makapili, and Maile soils had much larger PS0.2 values than the Kapaa, Haiku, and Halii soils, but the buffer coefficients of these soils suggested similar P sorption potential. The Wahiawa soil, with a medium sorption potential estimated with PS0.2, displayed a low sorption potential according to the buffer coefficients. These results suggest that predicting the P buffer coefficient should include other soil characteristics in addition to P sorption potential or P sorption site density.

Soil aggregation, as expressed by reactive mass and aggregate size, was correlated with the P buffer coefficient (Table 5) indicating that the prediction of the P buffer coefficients could be improved by including aggregate size and reactive mass. The best fitting statistical models quantify the effects of P sorption sites and reactive mass to the buffer coefficients (Table 7) . Signs of coefficients of each model indicated that the buffer coefficients were negatively related to both log of the sorption site density and reactive mass. Thus a soil with a low number of P sorption sites and a low fraction of reactive mass should have a large buffer coefficient, i.e., sorb relatively less P.


View this table:
[in this window]
[in a new window]
  		Table 7 Models and coefficients describing the prediction of P buffer coefficients of three soil tests for the incubations of 32 and 180 d

 
The models including clay mineralogy and physical factors substantially reduced standard errors of the buffer coefficients when compared with Cox's models (Cox, 1994), where the buffer coefficients were predicted only by the clay contents of soils (Table 8) . In case the P sorption site density is not available or very expensive to estimate, the inclusions of water content at matrix potential of -1.5 MPa (Jackman, 1997), the soil specific surface area (Jackman, 1997), or GMD, are other alternatives to improve the prediction of the P buffer coefficients (Table 8).


View this table:
[in this window]
[in a new window]
  		Table 8 The standard errors of the P buffer coefficient prediction by Cox's model (1994), the revised model in Table 7, and the other selected models

 
Although the prediction of buffer coefficients was improved by including P sorption density and soil aggregation, it still is necessary to test whether extractable P within aggregates has the same availability as extractable P on the surface of the aggregates. Current soil P test methods have not been evaluated for the effects of P diffusion into aggregates and accessibility of roots and root hairs to P within the aggregates on soil P supply.


   ACKNOWLEDGMENTS
 
This research was supported by the Soil Management-Collaborative Research Program of the U.S. Agency for International Development and the U.S. Department of Agriculture Section 406 Funds.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 Materials and methods
 Results and discussion
 REFERENCES
 
College of Tropical Agriculture and Human Resources Journal Series No. 4417.

Received for publication May 7, 1999.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Materials and methods
 Results and discussion
 REFERENCES
 




This article has been cited by other articles:


Home page
 Soil Sci.Home page
X. Wang, R.S. Yost, and B.A. Linquist
Soil Aggregate Size Affects Phosphorus Desorption from Highly Weathered Soils and Plant Growth
Soil Sci. Soc. Am. J., January 1, 2001; 65(1): 139 - 146.
[Abstract] [Full Text]

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 ISI 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 HighWire
Right arrow Citing Articles via ISI Web of Science (3)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Wang, X.
Right arrow Articles by Linquist, B.A.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Wang, X.
Right arrow Articles by Linquist, B.A.
Agricola
Right arrow Articles by Wang, X.
Right arrow Articles by Linquist, B.A.

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