SSSAJ Grow Your Career with SSSA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 20 September 2006
Published in Soil Sci Soc Am J 70:1825-1833 (2006)
DOI: 10.2136/sssaj2005.0285
© 2006 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 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 Google Scholar
Google Scholar
Right arrow Articles by Ruiz-Vera, V. M.
Right arrow Articles by Wu, L.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Ruiz-Vera, V. M.
Right arrow Articles by Wu, L.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Ruiz-Vera, V. M.
Right arrow Articles by Wu, L.
Related Collections
Right arrow Structure and Properties
Right arrow Soil Mineralogy
Right arrow Soil Salinity

Soil Physics

Influence of Sodicity, Clay Mineralogy, Prewetting Rate, and Their Interaction on Aggregate Stability

Victor M. Ruiz-Veraa and Laosheng Wub,*

a Colegio de Postgraduados, Campus San Luis Potosí, Iturbide 73, Salinas, S.L.P. 78600, Mexico
b Dep. of Environmental Sciences, Univ. of California, Riverside, CA 92521, USA

* Corresponding author (Laosheng.wu{at}ucr.edu)


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Sodicity adversely affects soil physical conditions reflected by weak structural stability. Soil clay mineralogy influences the degree of aggregate disruption induced by sodicity. The main objective of this research was to evaluate the interactive effects of clay mineralogy, prewetting rate (PWR), and sodicity on soil aggregation. Three soils with predominantly clay minerals of smectite, vermiculite-smectite, and kaolinite were equilibrated with NaCl–CaCl2 solutions having sodium adsorption ratio (SAR) values of 0, 20, and 50 and electrical conductivity (EC) = 3.0 dS m–1. After air-drying, the treated samples were packed and prewetted at rates of 2 and 30 mm h–1 with the NaCl–CaCl2 solutions. Aggregate-size distribution, macroscopic swelling, and surface soil dispersion were determined after the packed samples were equilibrated to a matric potential of –0.1 MPa. The kaolinitic soil showed the lowest inherent aggregate stability when subjected to slow PWR and the lowest SAR. Furthermore, aggregate stability of the kaolinitic loam soil was not significantly affected by increasing SAR. The Millox (smectitic) soil, on the other hand, was most susceptible to aggregate slaking, whereas the Malibu (vermiculitic) soil was most susceptible to differential swelling (at slow PWR). When SAR was low, aggregate slaking by fast PWR was the main cause of the aggregate breakdown. At SAR ≥ 20, swelling and dispersion became more important to the structural stability.

Abbreviations: EC, electrical conductivity • ESP, exchangeable sodium percentage • OM, organic matter • PWR, prewetting rate • SAR, sodium adsorption ratio • TEC, total electrolyte concentration


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
SOIL STRUCTURE ALTERATION is the primary soil response to an excess of exchangeable sodium in combination with low salinity, which results in a decline in soil air and water permeability (Oster and Shainberg, 2001) and consequently, diminishes agricultural soil productivity.

Quirk and Schofield (1955) identified the processes of slaking, dispersion, and swelling as the key factors involved in the soil structure degradation due to sodicity. Slaking is a physical process in which soil aggregates disintegrate, either by explosion of entrapped air or by differential swelling, into microaggregates or aggregates of smaller size than the original aggregates. When a dry soil aggregate is suddenly surrounded by liquid water, capillary and adsorptive forces drive the water into the aggregate, thereby compressing the entrapped air in the intra-aggregate pores. The entrapped air pressure increases in proportion to the compression until it exceeds the cohesive strength of the aggregate, which is often sufficient to disrupt the aggregates (Klute, 1986; Auerswald, 1995; Hillel, 1998) especially in non-swelling soils (e.g., kaolinic soils). Auerswald (1995) concluded that air entrapment was the main cause of aggregate disintegration of prewetted aggregates of 113 arable topsoils during percolation tests, while shear force of the percolating water, swelling, and clay dispersion had little or no effect on aggregate disintegration. Abu-Sharar et al. (1987) concluded that the extent of slaking depends on SAR and total electrolyte concentration (TEC). The slaking of a loamy soil occurred at SAR of 0, 10, and 20, and electrolyte concentrations of 3.2, 15.9, and 19.4 mol m–3, respectively, with the soil showing little or no dispersion. For lower TEC values, dispersion occurred during the final stages of the slaking process.

The swelling process, under sodic conditions, can be highly disruptive because ion hydration and osmotic swelling forces pull water into interlayer spaces between the clay platelets, thereby pushing clay particles apart and causing the breakdown of the aggregates of swelling soils. If wetting occurs quickly, the wetted portion of a soil can swell appreciably compared with the dry portion. This process, called differential swelling, causes the development of a shear plane on the wetting front, which can break many of the bonds between particles (Kemper and Rosenau, 1986).

Unlike swelling, soil dispersion is a non-reversible phenomenon. Spontaneous soil dispersion is favored by the presence of a high exchangeable sodium percentage (ESP) and a low salt concentration in the bulk solution, although soil dispersion can occur in soils with an ESP below 15 if the TEC in the bulk solution is very low. Other variables can affect the spontaneous dispersion; the most important ones include clay mineralogy, the clay charge, pH, organic matter (OM) content, and the presence of Fe- and Al-oxides.

Many physical and chemical properties and agricultural management practice can affect the slaking, swelling, and dispersion processes, and thus affect aggregate stability (Goldberg et al., 1988; Le Bissonnais and Arrouays, 1997; Boix-Fayos et al., 2001; Levy and Mamedov, 2002; Levy et al., 2003). However, poor correlation between aggregate stability and the soil physical and chemical properties found by different researchers suggests that aggregate stability is a complex function (Levy and Mamedov, 2002). For example, soil OM is considered the primordial cementing agent that has a strong influence on aggregate stability (Tisdall and Oades, 1982). Nevertheless, when the OM content is low (such as in soils from arid and semiarid regions), clay content, the amount of Fe- and Al-oxides, or calcium carbonate can govern aggregate stability. Boix-Fayos et al. (2001) found a positive correlation between water stability of microaggregates and clay content, whereas the stability of macroaggregates depended on the OM content only when the OM content was >5 or 6%. When the OM was <5%, aggregate stability was strongly affected by CaCO3 content. Keren and Ben-Hur (2003) also indicated that CaCO3, acting as a cementing agent, decreased the aggregate slaking of a Chromoxerert-sand mixture.

Levy et al. (2003, 2005) studied the combined effects of salinity, sodicity, wetting rate, and soil texture on the hydraulic conductivity and aggregate stability. They found that wetting rate has no effect on aggregate slaking of soils with low clay content (~9%). As clay content increases, there is a significant interaction among the treatment variables affecting hydraulic conductivity and aggregate stability. Tedeschi and Dell'Aquila (2005) used aggregate stability as an index to estimate the degradation of the soil physical properties caused by irrigation with saline water in a 7-yr experiment, and they found that aggregate stability decreases as ESP increases. They considered that high values of ESP caused the aggregate breakdown into elementary particles by physicochemical dispersion. Mamedov et al. (2001) and Shainberg et al. (2001) related the influence of soil texture and wetting to the responses of soils to salinity and sodicity, but they did not evaluate the influence of soil mineralogy and its interaction with salinity, sodicity, and water application rate.

In summary, the physical and hydraulic responses of soils to high ESP values and low salinity have been extensively studied. However, in contrast to soil texture, only a few studies included clay mineralogy (McNeal and Coleman, 1966; Emerson, 1977). Many soils with different clay mineral types have been used in individual studies, but few studies used several soils of different minerals to study their responses to a series of combined treatments of EC, SAR, and PWR. The objectives of this research were to (i) evaluate the interaction effect of sodicity, PWR, and clay mineralogy on aggregate stability and aggregate-size distribution; and (ii) assess the relationship between clay mineralogy and the role of TEC, SAR, and PWR on the relative importance of the processes of slaking, swelling, and dispersion on aggregate breakdown.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Soil samples were collected from one cultivated plot (Millox series; Bakersfield, CA) and two non-cultivated plots (Mokelumne series; Sacramento, CA and Malibu series; Oxnard, CA). The soils were selected based on their texture and clay mineralogy. Table 1 shows the taxonomic classification, major clay minerals (determined through X-ray diffraction analysis), and the basic physical and chemical properties of the test soils.


View this table:
[in this window]
[in a new window]
  		Table 1. Classification, major clay minerals, and selected physical and chemical properties of the three test soils.

 
Sodification
The soil samples were air-dried and passed through a 2-mm sieve. After sieving the soil samples were placed in series of 400 g (at approximate bulk density of 1.3 Mg m–3 for the Millox and Malibu soils, and 1.5 Mg m–3 for the Mokelumne soil) in Buchner funnels 18-cm i.d. and leached with chloride salts. The procedure consisted of leaching the soils first with a concentrated solution and then progressively reducing the leaching solution concentrations (Fig. 1 ). We achieved final SAR values of 0, 20, and 50; and EC = 3.0 ± 0.5 dS m–1 for the soil solutions at equilibrium.


Figure 1
View larger version (15K):
[in this window]
[in a new window]
  		Fig. 1. Brief description of the process (flow chart) used to increase the sodium adsorption ratio (SAR) of the test soils.

 
The detailed leaching process consisted of three steps (J.D. Oster, personal communication, 2003): (1) Leaching the soils with two pore volumes of a solution that has three times the desired SAR and contains sufficient Na to achieve the desired ESP (except for the Millox soil). The amount of Na+ needed for each soil was calculated from the CEC value (Table 1) and the desired ESP. Because exchanging Ca2+ with Na+ is not a chemically favored reaction, we used a percolating solution with an SAR equal to three times the desired ESP, a condition that favors the exchange of Ca2+ with Na+ (Table 2); (2) Leaching the soils with two pore volumes of solutions of 250 mol m–3 CaCl2 for SAR = 0; 231.66 mol m–3 NaCl and 134.17 mol m–3 CaCl2 for SAR = 20, and 382.70 mol m–3 NaCl and 58.61 mol m–3 CaCl2 for SAR = 50; and (3) Replacing the percolating solution subsequently with solutions of the respective SAR (0, 20, or 50), but with lower total chloride concentrations of 300, 100, 50, and 30 mmolc L–1 until achieving steady-state effluent SAR and EC values.


View this table:
[in this window]
[in a new window]
  		Table 2. Parameters used to calculate the composition of the initial leaching solution.{dagger}

 
The Millox soil (a calcareous soil) was leached with two to five pore volumes of a 500 mol m–3 NaCl solution and two pore volumes of a 300 mol m–3 NaCl solution to accelerate the exchange of Ca2+ by Na+ (in treatments with target SAR of 20 or 50). Step 3 was subsequently followed until a steady-state effluent of desired SAR and EC was achieved. The last effluent solution was collected and Ca2+, Mg2+, Na+, and K+ concentrations were determined using inducted coupled-plasma spectroscopy (ICP) to verify the final SAR values of the percolated solutions. After the leaching process, the samples were air-dried inside a greenhouse at ambient temperature between 20 and 30°C, crushed, passed through a 2-mm sieve, and stored in cardboard containers for approximately 3 mo.

Soil Packing
The air-dried <2-mm soil was packed to a bulk density ({rho}b) of 1.26 Mg m–3 for the Malibu, 1.28 Mg m–3 for the Millox and 1.51 Mg m–3 for the Mokelumne soils. The bulk densities of the packed cores were close to the respective undisturbed soils. The soil packing procedure consisted of placing soil in a brass cylinder (5.4 cm in diam.) comprised of three removable parts taped together: a lower ring of 1.4 cm in height, a middle ring of 6 cm in height and an upper ring of 2.3 cm in height. Enough soil was added and homogenized to fill the lower and middle rings, and the lowest 1 cm of the upper ring. After placing soil filled cylinder in a compacting machine, 50 to 250 strokes were applied, depending on soil type. After compaction, the lower and upper rings were removed and the middle-ring mass was determined. Only packed columns with the predefined bulk density (error margin ± 4.9 x 10–3 Mg m–3) were selected for later use.

Some of the samples previously sodificated and air-dried were sieved to obtain soil aggregates sized 1 to 2 mm for the determination of soil aggregate stability and aggregate-size distribution. A 1-cm high ring was filled with these soil aggregates at the same bulk density as the soil columns. The bottom section of the packed columns and rings were wrapped with double-layered cheesecloth to enable their subsequent use for the wetting experiments.

Wetting of Soil Samples
The purpose of wetting was to obtain samples with differences in aggregate disruption produced by different soil-wetting rates. The packed soil columns and the rings for this purpose were placed in a special wetting setup that allowed application of solutions of 3.00 mol m–3 CaCl2 for SAR = 0; 5.83 mol m–3 NaCl and 0.08 mol m–3 CaCl2 for SAR = 20; and 5.97 mol m–3 NaCl and 0.01 mol m–3 CaCl2 for SAR = 50. Samples previously equilibrated with solutions of a certain SAR (e.g., SAR = 20) and air-dried were subjected to a wetting treatment with a similar SAR solution (i.e., SAR = 20). The wetting rate was controlled by the means of a Mariotte device to deliver flow rates of 2 or 30 mm h–1. The Mariotte device consisted of a burette and a plastic tubing of 3-mm i.d. placed inside the burette to provide a constant hydraulic head. The burette was filled with the above mentioned CaCl2–NaCl solution of the proper SAR.

A piece of cheesecloth was placed on the surface of the soil to diminish the kinetic energy of the dripping solution. The samples were placed below the outlet of the Mariotte device resting on a plastic mesh that allowed free contact of the bottom with the atmosphere. When the waterfront reached the bottom of the soil column (at this point the soil was close to saturation), the solution application was terminated. The soil columns and the rings were then placed in a pressure chamber to achieve the water content equivalent to a soil water matric potential of –0.1 MPa.

Evaluation of Surface Soil Dispersion
The cheesecloth placed on top of the soil column was also used to obtain a qualitative measurement of the soil aggregate dispersion at soil surface. Quirk (1994) showed that soil particle dispersion starts when the total salt concentration of the percolating solution is below the turbidity concentration (Ctu), given by

Formula 1[1]
where Ctu is the electrolyte concentration (mmolc L–1) of the percolating solution at which dispersed clay particles appear in the percolate.

Once the soil surface reached saturation during wetting, some of the dispersed particles started to adhere to the cheesecloth fabric. When the wetting procedure was completed, the cheesecloth was removed and transferred to a 50-mL centrifuge tube. Twenty milliliters of distilled water were added to the centrifuge tube. The tube was shaken at three cycles per second for 1 h. Next, 6 mL of the suspension was transferred to a cuvette and the transmittance was determined at 420 nm using a spectrophotometer (Genesys 20, Thermo Spectronic). Samples with lower transmittance reflect higher surface-soil dispersion. We note that the cheesecloth method can be used only as a relative measure of dispersion for the same soil.

Aggregate Stability and Size Distribution
We used a modified method of Kemper and Rosenau (1986) and Elliott (1986) to determine aggregate stability of the soil samples previously subjected to different wetting rates. After equilibrium to a soil water matric potential of –0.1 MPa was reached, the ring with soil sample was transferred to a 2-mm sieve. This sieve was immersed for 5 min in an aluminum can filled with enough treatment solution to cover the soil sample. Next, the sieve was lowered and raised with a vertical displacement of 1.3 cm at 30 cycles per minute for 3 min. The fraction that passed through the 2-mm sieve was carefully transferred to a 1-mm sieve and the sieving process repeated. Again, the fraction that passed through the 1-mm sieve was transferred to a 0.25-mm sieve and the sieving procedure repeated.

To separate the sand fraction from the silt and clay fractions, the 2-mm sieve with the retained soil was immersed in 100 mL of dispersing solution containing 2 g L–1 of sodium hexametaphosphate. Sieving, as mentioned above, was performed for 3 min. The aggregates that remained stable after 3 min of sieving were gently rubbed across the screen with a rubber tipped rod. The same procedure was repeated for the retained fractions in the 1- and 0.25-mm sieves to determine the sand fractions in the respective aggregate-size fractions. All aggregate fractions were dried to 105°C for 24 h and the dry-weight of each fraction determined. Aggregate stability was calculated from the total mass of the aggregate fractions > 0.25 mm, divided by the sum of the weights of the three-size (1–2, 0.25–1, and <0.25 mm) fractions (Kemper and Rosenau, 1986). Although the smaller particles were subject to more sieving time, this extra time may not have a significant influence on the breakdown of aggregates. According to Kemper and Rosenau (1986), for the speed and displacement distance they used, the rate of aggregate breakdown is drastically reduced after 3 min of raising-lowering cycles. Therefore, the sieving cycle repetitions only serve as a standard method for aggregate separation.

Soil Swelling
Swelling (macroscopic swelling) of the soil column during the wetting procedure was measured at two soil matric potentials, 0 and –0.1 MPa. Soil swelling was determined from the difference in volume between the air-dried soil and the moist soil. The air-dried soil volume equaled the cylinder volume. We measured the swelling through the height increase. Horizontal swelling inside the column (which tends to reduce the pore space) was not evaluated in this research, although was considered indirectly by considering their effect on the infiltration rate (data not shown).

Analysis of variance and mean tests (P > 0.05) were performed on the aggregate-size distribution and surface soil dispersion data. The statistical variability of the soil swelling data was reported in standard deviation.


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Table 3 showed that the aggregate-size distribution and surface soil dispersion were affected by clay mineralogy, SAR, and PWR. A significant interaction among the three factors in their effect on aggregate-size distribution suggests that the susceptibility to aggregate slaking is complex and it depends on all the variables evaluated. Sodium adsorption ratio had a stronger influence than PWR on surface soil dispersion and on the breakdown of aggregates of size 1 to 2 mm into smaller aggregates (0.25–1.0 mm) and microaggregates (<0.25 mm) especially for the Malibu and Millox soils (Table 3). The breakdown of <0.25-mm microaggregates into aggregates of smaller diameter and primary particles was not evaluated in this study.


View this table:
[in this window]
[in a new window]
  		Table 3. Main effects of sodium adsorption ratio (SAR), prewetting rate (PWR), and clay mineralogy on aggregate size distribution and surface soil dispersion (evaluated as transmittance percentage).{dagger}

 
Other factors such as soil OM and clay content can also affect aggregate stability. However, the OM contents of the test soils, like other soils in an arid/semiarid region, are low (Table 1), and aggregate stability in such soils is not highly correlated with OM content (Goldberg et al., 1988; Le Bissonnais and Arrouays, 1997; Levy and Mamedov, 2002; and Levy et al., 2003), but positively correlated with clay content (Boix-Fayos et al., 2001; Levy et al., 2003).

Inherent Aggregate Stability
The slow PWR treatment with a solution of SAR = 0 reflects a low energy disruption of soils compared with fast prewetting with a higher SAR solution, in which the mechanisms for aggregate breakdown such as explosion of entrapped air and differential swelling may occur simultaneously (Le Bissonnais, 1996). Additionally, if TEC < Ctu, dispersion is expected to occur (Quirk and Schofield, 1955). The slow PWR treatment with a low SAR solution was considered the condition where inherent aggregate stability can be evaluated as a function of clay content (Levy et al., 2003) and clay mineralogy. In this condition neither explosion by entrapped air nor differential swelling are expected to produce aggregate breakdown. When the soil aggregate is wetted slowly, the pores inside an aggregate are filled with water with little air trapped inside them, which reduces the chance of explosion of entrapped air (Hillel, 1998; Auerswald, 1995). On the other hand, the slow PWR will reduce the differential swelling between moist and dry areas of an aggregate. Moreover, when SAR = 0 there is no macroscopic swelling.

The 1- to 2-mm aggregates was the dominant fraction after the Malibu and Millox soils were subjected to slow prewetting with a low SAR solution, while the Mokelumne soil showed a higher fraction of microaggregates at low SAR and slow prewetting. Only 28% of aggregates in the 1- to 2-mm fraction remained stable when the soil was leached using one pore volume of the 3 mol m–3 CaCl2 solution (SAR = 0) at a PWR of 2 mm h–1solution (SAR = 0) at a PWR of 2 mm h–1. This percentage was lower than the stable macroaggregates in the same size range of the Malibu and Millox soils (69 and 42% respectively). The remaining Mokelumne soil aggregates were disintegrated into the 0.25- to 1-mm (18%) and <0.25-mm (54%) fractions. The Malibu soil showed a higher inherent aggregate stability (at low PWR and low SAR). Twenty-one percent (g g–1) of the total soil aggregates were broken down to 0.25- to 1-mm aggregates; and only 10% to <0.25-mm microaggregates. The Millox soil aggregates had an intermediate breakdown rate in responding to the slow prewetting and low SAR treatment. Twenty-six percent of the total aggregates were in the range of 0.25- to 1-mm, while 32% of the aggregates were <0.25 mm (Fig. 2 Go –4 ).


Figure 2
View larger version (19K):
[in this window]
[in a new window]
  		Fig. 2. Effect of sodium adsorption ratio (SAR) and prewetting rate (PWR) on aggregate-size distribution of three soils ((a) Malibu, (b) Millox, and (c) Mokelumne) with different clay mineralogy. The figure shows the proportion of aggregates that remained on the group of size 1 to 2 mm after prewetting at 2 or 30 mm h–1 and sieved according to the aggregate stability procedure (the error bar indicates the standard error of the mean).

 

Figure 3
View larger version (20K):
[in this window]
[in a new window]
  		Fig. 3. Effect of sodium adsorption ratio (SAR) and prewetting rate (PWR) on aggregate-size distribution of three soils ((a) Malibu, (b) Millox, and (c) Mokelumne) with different clay mineralogy. The figure shows the proportion of aggregates of size 1 to 0.25 mm after prewetting at 2 or 30 mm h–1 with solutions of SAR 0, 20, and 50 and total chloride concentration of 6 mmolc L–1, and sieved according to the aggregate stability procedure (the error bar indicate the standard error of the mean).

 

Figure 4
View larger version (30K):
[in this window]
[in a new window]
  		Fig. 4. Effect of SAR and PWR on aggregate-size distribution of three soils ((a) Malibu, (b) Millox, and (c) Mokelumne) with different clay mineralogy. The figure shows the proportion of aggregates of size <0.25 mm after prewetting at 2 or 30 mm h–1 with solutions of SAR 0, 20, and 50 and total chloride concentration of 6 mmolc L–1, and sieved according to the aggregate stability procedure (the error bar indicate the standard error of the mean).

 
The lower inherent aggregate stability for the Mokelumne soil, even at slow prewetting, may be explained by its lower clay content compared with the other two soils. In addition, this kaolinitic soil has a much lower charge density than the Malibu and Millox soils, which can contribute to the weakening of the bonds when the Mokelumne aggregates are wet.

Influence of Prewetting Rate and Sodium Adsorption Ratio on Aggregate Slaking
As the PWR increased, the percentage of the 1- to 2-mm fraction decreased when the solution SAR was 0 for the Malibu soil or ≤20 for the Millox and Mokelumne soils. However, the effect of PWR at higher SAR was not significant for either the Malibu or the Millox soil (Fig. 2a and 2b). The increase in PWR did not affect the percentage of the 0.25- to 1-mm fraction (Fig. 3), but caused an increase in the percentage of <0.25-mm fraction at SAR ≤ 20 for the Millox soil and at SAR = 20 for the Mokelumne soil (Fig. 4b and 4c). More large aggregates (1–2 mm) breaking down to microaggregates at low SAR and fast prewetting indicate a higher susceptibility of these aggregates to slaking. A comparison in the increase of microaggregates (%) between PWR at 2 and at 30 mm h–1 (both at low SAR) showed that the Millox soil was the most susceptible to aggregate slaking (Fig. 4b). The microaggregate fraction for the Millox soil increased by 20% while the Malibu and Mokelumne soils increased by 7 and 3%, respectively, when PWR increased from 2 to 30 mm h–1. The Mokelumne soil has lower clay content than the other two soils. Levy et al. (2005) also reported a negligible aggregate slaking effect on the hydraulic conductivity of low clay content soil (~9%).

At the slow PWR, an increase in SAR from 0 to 20 and 50 significantly reduced the percentage of 1- to 2-mm aggregate fraction and increased the percentage of <0.25-mm fraction for the Malibu and Millox soils (Fig. 2 and 4). Again, a higher amount of microaggregates indicates a greater breakdown of 1- to 2- and 1- to 0.25-mm aggregates into microaggregates and thus, a higher susceptibility to sodicity. The Malibu soil showed the higher susceptibility to SAR increase. The microaggregate proportions increased by 76% for the Malibu soil and 64% for the Millox soil when SAR increased from 0 to 50 (at slow PWR) (Fig. 4a and 4b). The interaction effect SAR x PWR showed that the aggregate breakdown was dominated by sodicity for the Malibu and Millox soils. Swelling and dispersion due to high SAR values overshadowed the effect of PWR on aggregate slaking. This response is in agreement with the results of Levy et al. (2005) who observed that the clay swelling effect prevailed over the PWR on the decrease in hydraulic conductivity of clayey soils with ESP = 20.

Mechanisms for Aggregate Breakdown
The three mechanisms: entrapped air explosion, differential swelling, and clay dispersion (Le Bissonnais, 1996) that affect aggregate stability could all play a role in aggregate breakdown.

For the treatments with SAR = 0, the low Na and high Ca concentrations in the exchange phase of the clay particles and in the bulk soil solution prevented clay dispersion (Fig. 5 ). In addition, macroscopic swelling (given the SAR = 0) was unlikely to affect the aggregate status (Table 4). The absence of soil dispersion and the insignificant macroscopic swelling suggest that when PWR increased, aggregate disruption at this low SAR value was due to slaking by entrapped air explosion and differential crystalline swelling. The latter is a process that does not depend on the salt concentration as the macroscopic swelling does (Rengasamy and Sumner, 1998). It is important to note that crystalline swelling acted only on swelling soils, whereas explosion of entrapped air may have influenced the aggregate breakdown of all three test soils.


Figure 5
View larger version (14K):
[in this window]
[in a new window]
  		Fig. 5. Effect of SAR and PWR on surface soil dispersion (determined as transmittance [%] of suspensions obtained from dispersed surface soil) of the (a) Malibu (b) Millox, and (c) Mokelumne soils (bars indicate the standard error of the mean).

 

View this table:
[in this window]
[in a new window]
  		Table 4. Volume increase as a consequence of soil swelling during wetting of soil columns under sodic conditions.{dagger}

 
As PWR increases, an appreciable differential swelling occurs in between the moist and dry regions of an aggregate in a swelling soil. This process creates a shear plane at the moist-dry soil interface and causes the breakdown of the original soil aggregate into smaller aggregates (Kemper and Rosenau, 1986; Quirk and Murray, 1991). Abu-Sharar et al. (1987) suggested that when repulsion forces attained a level that allows only the shearing stresses to become operative, aggregate slaking can occur before clay dispersion. They considered that random breakdown of soil aggregates occurs at planes of weaknesses. The points of weakness are likely the coarse pores (Quirk and Panabokke, 1962; Quirk and Williams, 1974). Auerswald (1995) reported greater influence of air entrapment than shear forces of percolating water on the aggregate disintegration. Panabokke and Quirk (1957) also concluded that air entrapment is the major cause of aggregate disruption if a loamy soil is wetted at matric potentials less than –0.1 MPa. Le Bissonnais (1996) and Levy et al. (2003) mentioned that breakdown by entrapped air explosion decreases as clay content increases, whereas breakdown by differential swelling increases with increasing clay content. This conclusion is in agreement with our hypothesis that the aggregate breakdown in the Mokelumne soil was not likely to occur through differential swelling since the predominant clay mineral in this soil is kaolinite, a 1:1 nonexpanding clay mineral.

If PWR is kept slow and SAR is increased to 20, it is expected that entrapped air explosion will not occur, whereas swelling due to high SAR will affect aggregate breakdown on swelling soils (Malibu and Millox), but high SAR has no significant effect on Mokelumne soil (which is not affected by swelling). As both PWR and SAR increased, differential swelling was likely the dominant mechanism that caused the aggregate breakdown for the Malibu and Millox soils when SAR was ≤20; whereas soil dispersion became the dominant mechanism when SAR was 50 at any PWR. Emerson (1977) calculated crystalline swelling of smectite to be 25 times that of kaolinite. An expansion of the interlayer region of the 2:1 clay minerals reduces the pressure and decreases the likelihood of explosion of the entrapped air within the aggregate. If the pressure inside an aggregate falls below the atmospheric pressure, the air remains entrapped within the matrix. The opposite occurs in kaolinite (1:1 clay mineral) aggregates that do not swell and they will shatter as the air is released (Emerson, 1977). Both processes, crystalline swelling, and explosion of entrapped air (at fast PWR), can occur at low SAR and high salinity values. Nevertheless, it is expected that high sodicity and low salinity can reinforce the swelling effect (by macroscopic swelling) on aggregate breakdown. It is important to note that soil aggregate disintegration to smaller aggregates or even separate clay domains produced by slaking does not cause clay particle dispersion, and that low SAR values do not prevent slaking. Quirk and Panabokke (1962) did not observe slaking differences (described as incipient failure) when they used distilled water or a 1 M CaCl2 solution on aggregates wetted at a matric potential of –2 kPa.

Fast PWR does not produce further disintegration by clay dispersion (Rengasamy and Olsson, 1991). The lack of PWR influence on clay dispersion (Fig. 5) suggests that soil particles attached to the cheesecloth layer came from soil dispersion, not from aggregate slaking. Surface soil dispersion increased as SAR increased. This response agrees with Eq. [1]. For SAR = 50, the calculated turbidity concentration is 8.2 mmolc L–1, which is above the electrolyte concentration used in this research (6.0 mmolc L–1). Therefore, except for the kaolinitic soil, it is expected that spontaneous dispersion arose at this SAR value. Quirk (2001) mentioned that the structural organization disruption (clay domains or quasicrystals) at and below the turbidity concentration (Ctu) is the basic cause of soil permeability reduction to air and water.

Soil permeability reduction is caused by several processes. Slaking could take place in the Malibu and Millox soils at SAR 50 at the initial stages of fast PWR. Swelling and slaking processes reduced the pore size and therefore soil permeability to water (Keren and Ben-Hur, 2003). Quirk and Schofield (1955) concluded that extensive hydration, which occurs at high SAR and low TEC, promotes separation of quasicrystals and eventually partial blocking of soil pores. The same process likely occurred in our experiments for the soils subjected to leaching by a solution of SAR 50. When a zone (upper layer) of low permeability is created due to dispersion, water will move more slowly into the deeper layers because of the low permeability of the upper layer (Quirk 2001). Under this condition, aggregate slaking at the lower layer will not be significant since aggregate slaking decreases with reduction in PWR (Fig. 2Go–4) (Panabokke and Quirk, 1957; Quirk and Panabokke, 1962; Mamedov et al., 2001). Finally, a stage of increasing macroscopic swelling and a condition of TEC lower than Ctu can be established. At this point, clay particle dispersion causes the total disruption of conducting pores (capillary pores and micropores). The infiltration then stops and the permeability becomes near zero (Quirk and Schofield, 1955; Quirk 2001). The moist soil depth for a PWR of 30 mm h–1 and SAR 50 was only 2 cm in our experiment (plus 0.5 cm of vertical swelling) of the total column length of 6 cm. A deviation from this behavior was observed in the Mokelumne soil. Clearly, the turbidity concentration not only depends on clay content, SAR, and TEC, but also on clay mineralogy, as is suggested by the much lower dispersion of the Mokelumne soil than the Millox and Malibu soils at SAR 50 (Fig. 5). Suarez et al. (1984) observed similar response on two soils, Bonsall (smectitic) and Fallbrook (mixed). The decrease in optical transmission (inversely related to clay dispersion) of suspensions equilibrated with SAR 20 solutions started at an electrolyte concentration of 25 mmolc L–1 for the Bonsall (smectitic) soil, and at 15 mmolc L–1 for the Fallbrook soil.


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
The responses of aggregate stability and its size distribution to sodicity were further affected by clay mineralogy, clay content, and PWR. The kaolinitic (Mokelumne) soil had the lowest inherent aggregate stability. When the aggregates were subjected to nonsodic conditions at a low PWR, slaking, swelling, and dispersion were not a main factor affecting aggregate stability of the smectitic and vermiculitic soils. Instead, the higher clay contents of the two soils physically enhanced aggregate stability through the cementing action of the clay fraction on the soil matrix, compared with the kaolinitic soil that has the lowest clay content of the test soils. In this study, clay content should have a greater influence than OM content in the susceptibility to aggregate breakdown due to low OM content of the test soils.

The Millox soil was the most susceptible to aggregate slaking, whereas the Malibu soil was most susceptible to differential swelling (at slow PWR). Sodium adsorption ratio, TEC, PWR, and clay mineralogy can all affect aggregate breakdown and consequently aggregate stability. We conclude that SAR has a stronger effect on the soil aggregate deterioration than PWR for the Malibu (vermiculitic) and Millox (smectitic) soils. Nevertheless, slaking driven by rapid PWR is the dominant process that causes the kaolinitic soil aggregate disruption. The processes of dispersion, swelling, and slaking in smectitic (Millox) and vermiculitic (Malibu) soils are driven by sodicity and to a lesser extent by prewetting-caused slaking. As PWR and SAR increased, slaking by swelling during extensive hydration mostly determined the aggregate stability provided TEC was above Ctu. When TEC decreased to below Ctu, swelling and especially dispersion of quasicrystals in the clay fractions of the Malibu and Millox soils determined the aggregate breakdown.

Conceptually, aggregate breakdown can be attributed to different processes, but it is difficult to experimentally separate them since both aggregate slaking due to entrapped air and differential swelling can occur simultaneously.

Received for publication August 30, 2005.


   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 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 Google Scholar
Google Scholar
Right arrow Articles by Ruiz-Vera, V. M.
Right arrow Articles by Wu, L.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Ruiz-Vera, V. M.
Right arrow Articles by Wu, L.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Ruiz-Vera, V. M.
Right arrow Articles by Wu, L.
Related Collections
Right arrow Structure and Properties
Right arrow Soil Mineralogy
Right arrow Soil Salinity

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