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


     

Published online 27 October 2005
Published in Soil Sci Soc Am J 69:1912-1921 (2005)
DOI: 10.2136/sssaj2004.0400
© 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 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 (9)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Park, E.-J.
Right arrow Articles by Smucker, A. J. M.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Park, E.-J.
Right arrow Articles by Smucker, A. J. M.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Park, E.-J.
Right arrow Articles by Smucker, A. J. M.
Related Collections
Right arrow Soil Methods/Instrumentation
Right arrow Structure and Properties
Right arrow Soil Physics

Soil Physics

Erosive Strengths of Concentric Regions within Soil Macroaggregates

Eun-Jin Park* and Alvin J. M. Smucker

Dep. of Crop and Soil Sciences, Michigan State Univ., East Lansing, MI 48824

* Corresponding author (parkeun2{at}msu.edu)


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
The stability of soil aggregate structures is very important for controlling aggregate dynamics and associated biogeochemical soil processes that respond to management and other physical disturbances. We examined the mechanical strengths of concentric soil layers from the surfaces to the centers of individual soil macroaggregates and compared them with polar tensile strengths (Ts), total soil C contents, soil texture, and bulk density of aggregates. Aggregates were sampled at depths of 0 to 5 cm from conventionally tilled (CT), no tilled (NT), and native forest (NF) ecosystems of a Wooster (fine-loamy, mesic Typic Fragiudalf) and a Hoytville (fine, illitic, mesic Mollic Epiaqualf) soils. Erosive strength (Es) of aggregates was defined as the surrogate for erosive forces required to remove 1 g of soil during 1 min from the surface of a soil aggregate rotating along the abrasive wall of a soil aggregate erosion (SAE) chamber. Total Es values of macroaggregates were consistent with Ts of whole aggregates and were controlled by aggregate size and treatment. The Es increased with decreasing aggregate size and from the exterior to the interior regions of aggregates. Measured changes in soil C content, texture, and bulk density across the different regions within aggregates did not completely explain the spatial distributions of Es among concentric layers within macroaggregates. Higher contents of C and clay contributed to the greater strengths of the Hoytville soil aggregates, regardless of bulk density. On the other hand, Es of soil aggregates from coarser-textured Wooster soils were correlated primarily with bulk density and appeared to be independent of C content.

Abbreviations: CT, conventional tillage • Es, erosive strength • NF, native forest • NT, no tillage • SAE, soil aggregate erosion • Ts, polar tensile strength


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
THE STABILITY OF SOIL aggregate structure and associated pore arrangements are very important soil properties that control biogeochemical soil processes and whole aggregate responses to physical disturbances associated with soil management systems. Previous studies have reported the development of concentric spatial gradients and variabilities in several physical, chemical, and biological characteristics within soil aggregates due to the existences of roots, preferential soil solution flows, and drying/rewetting cycles surrounding aggregates. Park and Smucker (2005) reported greater porosities in external regions of macroaggregates from a silty clay loam soil and suggested pore expansion into the centers of aggregates may be associated with their stability during multiple drying–rewetting cycles. Macroaggregates (>250-µm diam.) are considered as a secondary soil structure associated with pore space, microbial habitat, and physical protection of organic matter (Carter, 2004; Christensen, 2001). The rearrangement of pores and particles within stable macroaggregates develops as clay minerals are pulled outward and deposited on outer surface layers of soil aggregates by increasing capillary forces during the drying process (Horn et al., 1995). Sextone et al. (1985) observed anaerobic centers and concentric gradients in O2 profiles within large aggregates (radius ≥ 10 mm). Aggregates provide spatially differentiated habitats for microorganisms, that is, external regions with more oxygen, nutrients, and coarse pores that expose microbes to predation and additional desiccation, and inner regions with less oxygen and small pores with greater protection (Hattori and Hattori, 1976). Horn et al. (1994) and Chenu et al. (2001) reported spatial distributions in microorganisms within aggregates and suggested these physical and chemical gradients depended on texture and porosity. These apparent relationships between spatially distributed properties and concurrent aggregate stabilities suggest additional unknown mechanisms of stabilization may be functional within aggregate interiors.

To date, these reported gradients within aggregates have been isolated by manual peeling (Priesack and Kisser-Priesack, 1993; Ghadiri and Rose, 1991; Chenu et al., 2001; Ilg et al., 2004). Santos et al. (1997) separated uniform concentric surface layers using a pneumatic SAE apparatus designed to remove the surface soil materials from aggregates rotating against an abrasive-walled cylinder and driven by an upward cyclone air flow pattern. To estimate the surface strengths of eroding aggregate surfaces, we improved the peeling method by Santos et al. (1997) by applying specific centrifugal forces on individual aggregates enclosed in a smaller SAE chamber. The smaller and enclosed SAE chambers enabled quantitative comparison of specific erosive strengths among multiple concentric layers within aggregates.

The strength of individual soil aggregates has been generally determined by polar tensile strengths (Ts) quantified by measuring polar-applied forces required to disrupt larger volumes of whole aggregates into smaller size fractions (Dexter and Kroesbergen, 1985). The Ts of soil aggregates are dependent on the types and the quantities of cations, dispersible clays, soil organic matter, and drying–rewetting cycles (Kay and Dexter, 1992; Barzegar et al., 1995; Rahimi et al., 2000). Drying–rewetting cycles often decrease the strength of aggregates by the disruptive forces associated with entrapped air and localized swelling within hydrating aggregates (Hussein and Adey, 1998; Kay et al., 1994; Kay and Dexter, 1992) More pores, microcracks, and failure zones cause a decrease in aggregate strength, whereas compaction and densification of soil increase strength (Grant et al., 2001; Munkholm and Kay, 2002). Soil organic matter is recognized as a cementing material contributing to aggregate stability, but its contributions to the strength of soil aggregates continue to be debated. Causarano (1993) reported that organic matter increases the strength of moist soil aggregates and decreases the strength of dry aggregates of clay soils. Imhoff et al. (2002) suggested that Ts of soil aggregates were positively related to clay and silt contents, poorly crystalline Fe oxides, and organic matter. Perfect et al. (1995) reported that organic matter increases strength at low sand contents but strength decreases with increasing organic matter at high sand contents.

The Ts is an index of the combined resistive forces within soil aggregates, yet it fails to directly evaluate the intraaggregate processes, which contribute to the physical, chemical, and biological gradients reported within aggregates. Therefore, we designed a new approach for measuring concentric distribution of Es within soil aggregates. The distribution of Es is determined by quantifying the erosive forces required to remove concentric layers from the surfaces and interiors of soil aggregates during their erosion into smaller aggregate diameters. The physical, chemical, and biological gradients within aggregates including microcracks, texture, C, cations, pore sizes, and biopolymers interactively influence the gradients of Es required to remove each aggregate layer. Contributions of each component to the strengths of whole aggregates has been emphasized, yet the spatial distributions and combined effects of texture, C, and porosity/bulk density within aggregates and their interactions with strength distribution within aggregates are essentially unknown.

The objectives of this study were to: 1) identify spatial distributions of Es within soil aggregates from different soil tillage management systems on two soil types and 2) compare total integrated Es with Ts and identify the separate and combined influences of soil texture, C, and bulk density on both Es and Ts with the goal of providing new and novel insights into the complexities of soil structure.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Soil Aggregate Samples
Soil aggregates were sampled from Wooster silt loam (Typic Fragiudalf) and Hoytville silty clay loam (Mollic Ochraqualf) soils subjected to long-term tillage treatments at the Wooster and Hoytville research sites at the Ohio Agricultural Research and Development Center in October 1998. Nondisturbed soil blocks (15 x 15 cm) were extracted using a sharp flat spade from the surface 0 to 5 cm soils of conventionally tilled (CT) and no-tilled (NT) continuous corn experiments established at Wooster in 1962 and Hoytville in 1963. Samples were also extracted from the same soil types in adjoining native forests (NF) containing mostly deciduous trees. Soil samples were transported to the laboratory in rigid and sealed plastic containers. During air drying, soils were manually broken into smaller than 20-mm aggregates along their natural planes of weakness by applying gentle manual forces to the large soil aggregates. Care was taken to exclude the side portions of the original soil cube, which may have been compressed during the sampling process. Air-dried soils were manually sieved for <1 min to obtain the distribution of soil aggregates <1, 1 to 2, 2 to 4, 4 to 6.3, 6.3 to 9.5, and >9.5 mm and each size fraction was stored in rigid plastic containers at room temperature. Aggregate fractions used for this study, were 4 to 6.3 and 6.3 to 9.5 mm and represented 31 and 24% of bulk soils, by weight, for the Hoytville and Wooster soils.

Aggregates were peeled using SAE chambers (Fig. 1A) to sequentially remove 1/3 and 2/3 by weight of whole aggregates and retain their 1/3 exterior and 1/3 interior regions (Fig. 1B). Following the removal of exterior 1/3 region of the whole aggregate, the peeled materials retained in the base was collected and the remaining 2/3 portion of aggregate was transferred to cleaned SAE chamber for further peeling of transitional region. Separated exterior 1/3 and interior 1/3 region samples were saved for total C and texture determination. A detailed description of the peeling protocols by the SAE chambers are described further in next subsection.



View larger version (38K):
[in this window]
[in a new window]
  		Fig. 1. (A) The diagram of SAE chamber assembly system that includes a top erosion chamber containing knurled walls and a screen (350 µm) base, and a lower retainer chamber, which collects eroded soil materials. (B) Erosive strength were calculated for numerous concentric layers eroded from soil aggregates by abrasive forces exerted at the knurled wall surfaces of the SAE erosion chambers and C and texture were determined for exterior region and interior region samples collected when percentage of peeled mass reached 33.3 ± 2 and 66.7 ± 2%, respectively.

 
Identification of Erosive Strength within Macroaggregates
We quantified the erosive forces exerted on the surface of soil aggregates as the contiguous surface layers of each aggregate were peeled in the SAE chamber assembly system (Fig. 1A). When the external erosive forces applied to the surface of each aggregate are greater than the internal resistant strength of each soil layer, then soil materials on surfaces of the rotating aggregate are eroded and the rate of erosion is proportional to the difference between external erosive forces and internal resistances referred to as the Es. Therefore, we defined Es as the surrogate for erosive forces required to remove 1 g of soil during 1 min from the surface of a soil aggregate rotating along the abrasive wall within each SAE chamber.

Each stainless steel SAE chamber has a diameter of 25.4 mm and depth of 30.0 mm (Fig. 1A). The entire wall portion inside the erosion chamber and above the base screen was precisely machined into a uniformly knurled surface. Knurling was at 1 mm intervals, 0.4 mm deep, and at a 30° angle from perpendicular, using a circular pitch knurling tool (EPL-230, Form Roll Die Corp., Worcester, MA). A screen with openings of 350 µm, was welded to the base of the SAE erosion chambers. A stainless steel base was precisely machined to clearances of <0.05 mm to fit over the outside of the SAE erosion chamber and below the screen to retain all eroded materials, which passed through the screen.

At the onset of aggregate peeling, the erosion chamber and base was assembled, a pre-weighed single aggregate was placed on the screen of the SAE chamber and the top of the SAE chamber covered with heavy duty aluminum foil. The enclosed SAE chamber assembly was fitted into a foam envelope which tightly fit into 150-mL beakers and placed into the spring mounts on a rotary shaker platform containing up to 45 mounts (Innova, Model 2300, New Brunswick Scientific Inc., Edison, NJ). The Innova rotary shaker was equipped with a timer, tachometer, and variable speed controller that ranged from 0 to 500 rpm. The SAE chambers were firmly secured onto the shaker platform so that they did not move during the time when the shaker platform was rotating. Groups of SAE chamber assemblies, containing individual aggregates were rotated at speeds ranging from 200 to 400 rpm.

Frictional forces were generated at the surface interface of each aggregate, rolling clockwise along the wall of the SAE erosion chamber by the inertial momentum caused by counterclockwise rotational motion of the SAE chamber. At the onset of aggregate erosion (first few minutes at slow rpm), the sharp polygon surfaces of most natural aggregates were eroded by both the knurled wall and the screen at the base of erosion chamber. Aggregates that broke into multiple units, during the initial "polishing" treatment, were discarded. Natural soil aggregates became spherical after a few minutes of rotation and the aggregate surface maintained continuous contact with the knurled wall of the SAE erosion chamber. When each soil aggregate became more spherical, the rpm speed was increased applying greater centrifugal forces to the aggregate surface. Increased rpm lifted aggregates above the screen and aggregate erosion occurred primarily along the knurled wall of the SAE erosion chamber.

Thirty or more SAE chamber assemblies, each containing one aggregate, were rotated on the rotary shaker for 10 min. Each soil aggregate remaining on the screen was weighed to determine the rate of erosion. The time required to remove two concentric regions, each 1/3 of the original aggregate weight was calculated assuming constant erosion rate (Fig. 1B). As the erosion rate generally decreased with peeling aggregate, the aggregate was weighed 1 to 3 times before the first 1/3 (exterior) region was separated and another 2 to 6 more times before the second 1/3 (transitional) region was removed. The soil masses and erosion rates at these measurement times were used to calculate Es for each aggregate. Different intervals of time for weighing the aggregate, which remained on the SAE screen were used for each group of aggregates with similar erosion rate. Initial rotational speeds were 200 rpm for Wooster soils and 250 rpm for Hoytville soils. These were increased by 50 rpm up to 300 and 400 rpm, respectively, to provide the centrifugal erosive forces required to remove the more secure aggregate surfaces within aggregate interiors, without breaking the remaining aggregate whose diameters were diminishing. The speeds applied to each soil were determined by experience and consideration of aggregate strengths. Generally, the speed was increased after the first 1/3 (exterior) region was removed and 1 or 2 additional times for aggregates with very slow erosion rates while removing the next 1/3 (transitional) region.

When the SAE chamber assembly is rotating in a specific orbital radius at a speed controlled by the rotary shaker, we can describe the circular motion as diagrammed in Fig. 2 . The rotary shaker exerted a counterclockwise orbital radius of 12.50 mm. Contact points between the aggregate and the knurled wall form a circular track (S1) as shown in Fig. 2. The centrifugal force (C), with units of g mm s–2 = 10–6 Newton (N), exerted at the aggregate surface in each SAE erosion chamber can be calculated as follows:

[1]
where m(t) is the mass (g) and Ra(t) is the radius (mm) of each aggregate that changes with time t (s) as each aggregate is eroded. Radius (Ra) of eroded spherical aggregate at time t was calculated using aggregate mass measured at time t and different bulk density values for exterior, transitional, and interior regions which were determined in a previous study (Park and Smucker, 2005). The RSAE is the radius of erosion chamber (12.7 mm), and {omega} is the angular velocity (rad s–1) determined by the rotational speed of the rotary shaker (rpm) as follows:

[2]



View larger version (41K):
[in this window]
[in a new window]
  		Fig. 2. Diagram of the rotational clockwise motion of a soil aggregate as the surface is randomly abraded along the wall of the SAE chamber. The SAE chamber is stationary and moving along a rotary pattern on the platform of the orbital shaker. Positions p1, p2, p3, ... pn represent the rotational clockwise pattern of the aggregate at times t1, t2, t3, ... tn.

 
The actual friction force exerted at the aggregate surface is proportional to the centrifugal force C and frictional coefficient, which is determined by roughnesses of the knurled wall and the aggregate surface. Assuming uniform roughness of the erosion chamber walls of the SAE assembly and uniform roughness of eroded aggregate surfaces after sharp edges have been removed during the first few minutes of erosion, the frictional coefficient was considered to remain constant for the duration of the applied rotational force. Incorporating both these measured and observed assumptions, we could substitute the calculated Es to estimate the forces required to remove soil materials from surfaces of soil aggregates eroded within each SAE assembly. The comparable forces applied to the soil aggregate surface were calculated using the mass of soil removed during a given time of applied centrifugal force:

[3]
where m(tn) and m(tn–1) are the masses of soil aggregate materials removed at times tn and tn–1, respectively. The Es of 4 to 10 concentric layers within each soil aggregate were determined during the removal of the external 2/3 regions of individual aggregates using Eq. [1] through [3]. The numbers of Es calculations for concentric layers identified within each aggregate was controlled by their erosion rates. Slower aggregate erosion rates required longer peeling times, generating more measurements. Concentric layers from the exterior to interior regions of aggregates were expressed by the percentage of soil mass removed from aggregate surface by peeling (% peeled mass). The number of aggregates eroded for Es determinations for each treatment ranged from 24 to 86. This range resulted from a consolidation of time savings and the variability of aggregate samples. A commercial firebrick standard was used to further test the use of SAE chamber system for determining Es values. Firebrick fragments were broken, sieved into fractions 4 to 6.3 mm and rough edges of 15 samples were removed by peeling for a few minutes in the SAE chamber. Then the more spherical samples were further peeled as described above. These firebrick standards had an average bulk density of 0.75 g cm–3. Erosive strength values of the commercial firebrick standard were compared with the natural soil aggregate samples.

Total Carbon and Texture Analyses
Total C was determined for crushed whole aggregate samples from both aggregate size fractions and peeled exterior 1/3 and interior 1/3 region samples from the 6.3- to 9.5-mm aggregate-size fraction. Total C was determined for the sand-free samples by the dry combustion method using a Carlo-Erba, model NA1500, series 2 Nitrogen-Carbon-Sulfur Analyzer (Carlo-Erba, Milan, Italy). Sand-free samples were obtained by sieving with 53-µm sieve after gently grinding the soil samples.

Textures were also determined for whole aggregate samples from both aggregate-size fractions and for 1/3 exterior and 1/3 interior regions peeled from the 6.3- to 9.5-mm size fraction. Carbonates were removed from 2 g of crushed or peeled aggregate samples using sodium acetate (pH 5), soil organic matter was removed by heating with NaOCl (pH 9.5), and Fe oxides were removed with sodium citrate and dithionite as described by Kunze and Dixon (1986). After carbonates and organic matter were removed, the soil samples were dispersed with sodium hexametaphosphate (5 g L–1) before clay contents were determined by Stokes's law using a modified syringe method described by Moshrefi (1993). Forty five milliliters of dispersed soil suspension were transferred into a 50-mL sedimentation cylinder having three septa at different distances from the solution surface. Five milliliter of clay suspension were collected through each septum using a syringe at calculated sedimentation times and dried at 105°C. Total clay contents were calculated from the clay contents of these subsamples. Sand contents were determined after washing the dispersed samples through 53-µm sieve. Silt contents were calculated by difference of the measured clay and sand. We used the bulk density values for the exterior and interior regions of aggregates and whole aggregates determined for the same soil samples reported in a previous study (Park and Smucker, 2005).

Polar Tensile Strength of Whole Aggregates (Ts)
The air-dried aggregates were crushed between a bottom flat plate placed on an electronic digital balance and a top flat plate connected to a computer-controlled stepper motor. The largest surface plane of each aggregate was placed on the bottom plate before vertical compressive force was applied by the top plate. The increasing compressive forces were continuously recorded by computer and the forces were identified when aggregates were fractured. Thirty aggregates of each size fraction from each treatment and soil type were crushed. Polar tensile strengths having units of N m–2 or Pascal (Pa) were calculated by the equation (Dexter and Kroesbergen, 1985):

[4]
where F is the crushing force required to fracture the aggregate and determined by

[5]
where b is the measured balance reading (kg) and g is the gravity acceleration (9.807 ms–1). The effective diameter of each aggregate, d, was calculated as d = dm (m/m0)1/3, where dm is the mean of the openings of upper and lower sieves used for aggregate-size fractionation, m and m0 are the mass of individual aggregate and the mean mass of replicates, respectively.

Statistical Analyses and Integration of Erosive Strength
Data were analyzed for a completely randomized block design using the SAS statistical package (SAS Institute, Cary, NC). The effects of tillage treatment and aggregate size were tested at significance level of P < 0.05 using PROC MIXED in SAS/STAT. SAS PROCNLIN in SAS/STAT regression analysis between Es and percentage of peeled mass was conducted on each aggregate to interpolate discrete Es values and to take the average Es for percentage of peeled mass of the 24 to 86 replications from each management and soil type. The relations between Es and percentage peeled mass were best compared by an exponential equation (P < 0.05 for all aggregates tested):

[6]
where X is the location of the concentric layer within the aggregates, expressed as percentage of peeled mass from the exterior to the interior regions within aggregates, a and b are the regression constants of the exponential equation. Differences among constants a and b were analyzed to identify significant differences in Es among treatments and aggregate sizes at P < 0.05 using PROC MIXED with Tukey adjustments. The values of Es from 0 to 33.3, 66.7 to 100, and 0 to 100% were integrated using the integration form of Es, {int}aebXdx to calculate total Es for exterior 1/3 regions, interior 1/3 regions, and whole aggregates.


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Peeling Time of Soil Aggregates
The erosion times required to peel multiple concentric layers from 0 to 66.7% of the original mass of soil aggregates, using the SAE chamber systems, were uniquely different when comparing soil types, soil aggregate-size classes, and field management treatments (Fig. 3) . For instance, there was approximately 100-fold difference in erosion times required to remove exterior 2/3 regions of NF aggregates between Hoytville and Wooster soils, that is, about 3000 min (Fig. 3A) vs. 30 min (Fig. 3B). Erosion time differences between the two soil types would be even greater if both were eroded at the same rpm speed. Greater erosion times were required for smaller 4.0- to 6.3-mm aggregates of the same management treatments in the Hoytville soil (Fig. 3A). Among the three management groups, the erosion rates (the slopes of the curves) for Hoytville soil aggregates were CT > NT > NF (Fig. 3A), whereas Wooster aggregates from both CT and NT treatments were eroded at similar rates that were lower than those of Wooster NF aggregates (Fig. 3B).



View larger version (23K):
[in this window]
[in a new window]
  		Fig. 3. The average measured percentages (n = 10) of soil masses removed from aggregate surfaces with increasing erosion times for different aggregate-size fractions from field CT, NT, and NF management systems on (A) Hoytville silty clay loam and (B) Wooster silt loam soils. Some percentage of peeled values not measured at the given erosion time were linearly interpolated between two closest points to take the average of 10 aggregates. Exterior, Transitional, and Interior represent concentric regions within aggregates as presented in Fig. 1B.

 
Decreasing aggregate erosion rates over time for all treatments and aggregate-size fractions can be observed by the slopes for each measurement recorded in Fig. 3. These declines can be explained by either a reduction in the soil aggregate mass during the peeling, or by increased strengths of layers closer to the center of the aggregates. Therefore, changes in the centrifugal forces associated with diminishing aggregate mass and increasing rotary shaker speeds needed to be standardized before it would be possible to compare specific erosive strengths among concentric layers within soil aggregates from different soil types and management systems. Erosion rates under standardized constant centrifugal forces can be evaluated. Or, inversely, centrifugal forces required to maintain constant erosion rates during the peeling of aggregates, could become a meaningful parameter that would improve our understanding of how internal aggregate strengths contribute to various aggregate functions.

Erosive Strengths within Concentric Layers of Soil Aggregates
The Es values within soil aggregates from Hoytville and Wooster soil series were lowest at the surfaces and increased toward the centers of the aggregates (Fig. 4 and 5) . These concentric gradients in Es within Hoytville and Wooster soil aggregates were contrasted to no significant gradient in Es within firebrick standard samples. The average Es values from exterior to interior layers within the commercial firebrick samples ranged from 1.5 x 105 to 1.8 x 105 N (kg s–1)–1 and were greater than those of Wooster soil aggregates and lower than the Hoytville soil series. Aggregates from NF treatments of the Hoytville soils exhibited the greatest Es that were approximately 200-fold greater than aggregates from the NF treatments of the Wooster soils. Soil aggregate erosive strengths decreased in the order of NF > NT > CT for Hoytville soils and were reversed as CT > NT > NF in Wooster soils. However, the values of constants a and b of a regression Eq. [6] shown in Table 1, indicate there were no significant differences among the Es values of intraaggregate regions within CT and NT aggregates from Hoytville. The reversed effects of management treatment on erosive tensile strength values for aggregates from Hoytville and Wooster soil series were attributed to the interactive influences of soil C and clay contents, and bulk density on Es, described in greater detail in subsequent subsection.



View larger version (20K):
[in this window]
[in a new window]
  		Fig. 4. Cumulative increases in erosive strength (Es) in aggregates, (A) 6.3 to 9.5 mm and (B) 4 to 6.3 mm, from Hoytville silty clay loam soils at depths from 0 to 5 cm, as aggregates were peeled in SAE chambers. These Es lines were obtained by taking average of Es values fitted in a regression Eq. [6] for each aggregate (Table 1). R2 values represent a measure of closeness between measured data and fitted lines for each treatment.

 


View larger version (22K):
[in this window]
[in a new window]
  		Fig. 5. Cumulative increases in erosive strength (Es) in aggregates, (A) 6.3 to 9.5 mm and (B) 4 to 6.3 mm, from Wooster silty loam soils at depths from 0 to 5 cm, as aggregates were peeled in SAE chambers. These Es lines were obtained by taking average of Es values fitted in a regression Eq. [6] for each aggregate (Table 1). R2 values represent a measure of closeness between measured data and fitted lines.

 

View this table:
[in this window]
[in a new window]
  		Table 1. The means and standard errors of constants, a and b in erosive strength (Es) = aebX (Eq. [6]), fitted relation between Es and percentage of peeled (X) for aggregates from different managements of two soil types.

 
The Es of soil aggregates increased with decreasing aggregate size for all management treatments of the Hoytville soil (Fig. 4 and Table 1). However, differences in Es between the two aggregate-size fractions were not significant for the NT and CT aggregates from Wooster soils (Fig. 5 and Table 1). The Es of NF aggregates from both soil types responded more to changing aggregate-size fractions. Causarano (1993) reported that the Ts of soil aggregates increased by 50% with decreasing aggregate size from 30 to 10 mm and suggested that less surface cracks and imperfections and a more dense structure contributed to greater Ts of smaller aggregates.

The gradients of Es within soil aggregates appear to be positively correlated with the larger porosities at aggregate surfaces and lower porosities in the centers of aggregates (Park and Smucker, 2005). The outer surface of aggregates is more subjected to harsh drying stresses and high energy input associated with clay shrinkage, which lead to greater microcracks and weakest links that are preferentially removed during peeling. This is also apparent when different size aggregates are compared. The larger aggregates can be eroded more easily because the lower ratio of surface area to volume resulted in more surface cracks during drying process.

Relationships between Erosive Strength and Tensile Strength
Aggregate size and management treatment modifications of Ts and integrated Es from 0 to 100% peeled were highly correlated (R2 = 0.98, P < 0.001 for Hoytville soils and R2 = 0.83, P = 0.039 for Wooster soils). Tensile strength increased with decreasing aggregate size for both soil series of this study agrees with Causarano (1993). Aggregate size reductions, from 6.3 to 9.5 to 4 to 6.3 mm, caused Ts to increase by 39, 34, and 106% for CT, NT, and NF in Hoytville and 81, 51, and 83% in Wooster, respectively (Fig. 6) . There were no significant differences in Ts between NT and NF for the larger Hoytville aggregates or between CT and NT in the larger Wooster aggregates. However, Ts of smaller aggregates (4–6.3 mm) increased significantly in the order of NF > NT > CT for Hoytville soils while CT > NT > NF for Wooster soils and these trends in Ts were consistent with those in Es of aggregates of the same size fraction for each of the different soil types. These high correlations between Ts and Es indicate that the Ts of an aggregate is controlled by the sum of concentric layer tensile strengths resisting erosion within soil aggregates or the accumulation of the internal tensile strengths at the concentric layers within that aggregate. The Ts of aggregates appear to be more related to the connections of major failure points and cracks throughout the aggregate leading to the breakage of the whole aggregate (Horn et al., 1995). As measured in this study, Es appears to be a combination of microsite tensile strengths exhibited by individual regions removed by applications of uniform erosive energies. These milder applications of energy, by the SAE chambers, erode aggregate surfaces without destroying the remaining whole aggregate, providing opportunities to identify gradients of C, ions, texture, bulk density and porosity which control aggregate tensile strengths.



View larger version (23K):
[in this window]
[in a new window]
  		Fig. 6. Tensile strength (Ts) of whole aggregates from (A) Hoytville silty clay loam and (B) Wooster silt loam soils sampled at depths from 0 to 5 cm. Bars represent the standard error of three field replicates (n = 30 for each field replicate).

 
Relations between Aggregate Strengths and Carbon, Texture, or Bulk Density
According to the hierarchical concept of aggregate formation processes, larger aggregates contain more macropores than the smaller and denser aggregates within them (Dexter, 1988). Consequently, macroaggregates composed of microaggregates can be more easily fractured along the larger pores and cracks. Textures and C contents were similar between the two aggregate-size fractions of each soil type but there was a tendency that bulk density of soil aggregates increased with decreasing aggregate size for all soils although it is not statistically significant in Wooster soils (Table 2). These changes in bulk density with decreasing aggregate size in two soils were consistent with the changes in Es with decreasing aggregate sizes. Therefore, increases in aggregate strengths with decreasing aggregate size likely results from greater cohesive forces associated with increased bulk density.


View this table:
[in this window]
[in a new window]
  		Table 2. Selected physical and chemical properties of two aggregate size fractions from a Hoytville silty clay loam and Wooster silt loam, sampled at depths from 0 to 5 cm (n = 3).{dagger}

 
However, increased bulk densities of soil aggregates, caused by cultivation and tillage did not consistently increase aggregate strengths for both soil types (Table 3). The order of bulk density is CT > NT > NF in both soil types (Table 2), whereas aggregate Es and Ts strengths were parallel to bulk density for Wooster and inversely related for Hoytville (Fig. 4, 5, and 6). The strength of dry aggregates appears to be influenced not only by bulk density but also by C content, texture, and their interactions. The interactions among C, texture, and porosity bring a different role of organic C in tensile strength depending on texture. We observed that C contributed to strengthening aggregates in more clayey soils from Hoytville but aggregate strengths decreased with increasing C content in coarser-textured soils from Wooster (Table 3), which corresponds to a result reported by Perfect et al. (1995). These observations support the interactions among C, texture, and porosity suggested by Kay (1998). He explained that increasing organic C causes a greater increase in the porosity of coarse textured soils than fine textured soils and contributes to the loss of soil strength, whereas more organic C more effectively increases bonding between minerals and increases strength where the matrix is dominated by clay. However, the biogeochemical mechanisms causing the greatest aggregate strengths to occur in the NF treatments of the finer textured Hoytville soil aggregates (Table 2 and Fig. 4 and 6) remains unclear.


View this table:
[in this window]
[in a new window]
  		Table 3. Pearson correlation coefficients between Ts and other properties of soil aggregates for different size fraction and soil type (n = 9). The values in parentheses are the statistical significances of coefficients.

 
The values of C, bulk density, and Es corresponding to the exterior and interior of aggregates are compared in Table 4. Neither the C gradients nor the changes in bulk density directly explain the significant increases in Es from exterior to interior regions of soil aggregates. We also found no differences in texture between these two regions for both soil types (data not shown). It appears that soil C, bulk density, and texture are not linearly related to aggregate strengths for both Hoytville silty clay loam and Wooster silt loam soils, as quantitative considerations of these factors do not completely explain the major contributors to spatial distributions of soil strengths within aggregates. Aggregate density measurements are probably not sensitive enough to identify the subtle changes in microcrack density that would have a large influence on Es.


View this table:
[in this window]
[in a new window]
  		Table 4. Exterior to interior gradients in C, bulk density ({rho}b), and erosive strengths (Es) within soil aggregates, 6.3 to 9.5 mm, from Hoytville silty clay loam and Wooster silt loam soils sampled from depths of 0 to 5 cm.

 
Limitations and Implications for Future Studies
There may be a limit in the determination of Es values for soil aggregates with extraordinarily low stability due to the heterogeneous cementation of aggregate particles, especially those having significant quantities of coarse sand. However, our observations and experiences with the peeling of literally thousands of aggregates for soil textures ranging from loamy sand to clay confirm that aggregates from a broad range of soil types can be effectively peeled by the SAE chambers. The equations can be applied to the data generated by the SAE chambers to further quantify the Es values of specific concentric layers within soil macroaggregates.

The assumptions on the shape and the surface roughness of soil aggregates are associated with major limitations of this study. Erosive strength was determined for ideal spherical shape and constant surface roughness of aggregates. The great differences in Es among soil samples and concentric layers reported in this study should be ascertained with the effects of the changes in shape and surface roughness during peeling. In particular, the exponential relationship between Es and percentage of peeled mass could be more related to the attrition of rough edges of surface aggregates, that is, rounding aggregates, than the spatial distributions of binding agents and cracks. It can be clarified using ideal mixes such as sand mixed with different concentrations of binding agent controlling strength (Mullier et al., 1991). The approach also needs further testing for soil samples with various shapes, sizes, and spatial distributions of particle sizes, pores, cracks, and cementing agents. We tested two size fractions of three managements and two soil types that demonstrate combinations of these components. The observed differences in Es among aggregate samples and layers should be further investigated with the contributing factors in microscale revealing the subtle changes, especially, more accurate measurements of pore arrangement and the spatial contributions of cementing agents.


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
We developed a novel approach for identifying intraaggregate strengths from exterior layers to interior layers of aggregates by the mechanical abrasion of soil aggregates enclosed within SAE chambers. Erosive strength, a surrogate for resistance to erosion was defined as the force required to erode a unit weight of soil material from the surface of individual aggregates during a period of time by continuous and uniform abrasion along the knurled walls of manufactured stainless steel SAE chambers. Summations of concentric layer Es values were consistent with Ts of whole aggregates and changed with aggregate size and soil management practices. The Es of soil aggregates increased with decreasing aggregate size and increased from exterior layers to interior regions. We concluded that stronger central regions of soil aggregates are more protected and may turn over less frequently than aggregate exteriors. There was a mixed response of spatial Es differences within aggregates having contrasting soil C contents, bulk densities, or textural differences. Evaluations of the spatial distribution of soil strength within aggregates, initiated by the application of the SAE approach used in this study, offer new opportunities to expand our understanding of the complex interrelationships among soil minerals, soil C and bulk density/porosity and texture. The SAE peeling of concentric layers also provides additional understanding of aggregation processes related to the spatial gradients of biogeochemical components within aggregates.


   ACKNOWLEDGMENTS
 
Financial support was provided by USDA/CSREES Projects Nos. 2001-38700-11092 and S03057 (CASMGS). We are grateful to Dr. Alexandra Kravchenko for advice on statistical analyses and Karin Adtjandra for laboratory assistance. Soil samples provided by Warren Dick, Ohio State University, OARDC, from Hoytville and Wooster, OH, are greatly appreciated. We also thank Peak Industries for fabrication of SAE chamber assemblies.

Received for publication December 22, 2004.


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




This article has been cited by other articles:


Home page
 Vadose Zone JHome page
A. J.M. Smucker, E.-J. Park, J. Dorner, and R. Horn
Soil Micropore Development and Contributions to Soluble Carbon Transport within Macroaggregates
Vadose Zone J., May 17, 2007; 6(2): 282 - 290.
[Abstract] [Full Text] [PDF]

Home page
 Soil Sci.Home page
J. Kinyangi, D. Solomon, B. Liang, M. Lerotic, S. Wirick, and J. Lehmann
Nanoscale Biogeocomplexity of the Organomineral Assemblage in Soil: Application of STXM Microscopy and C 1s-NEXAFS Spectroscopy
Soil Sci. Soc. Am. J., August 22, 2006; 70(5): 1708 - 1718.
[Abstract] [Full Text] [PDF]

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 (9)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Park, E.-J.
Right arrow Articles by Smucker, A. J. M.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Park, E.-J.
Right arrow Articles by Smucker, A. J. M.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Park, E.-J.
Right arrow Articles by Smucker, A. J. M.
Related Collections
Right arrow Soil Methods/Instrumentation
Right arrow Structure and Properties
Right arrow Soil Physics

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