Comparison of four methods for liberating various aggregate fractions in Vertosols to study their morphology.
The presence of stable aggregates at the soil surface affects the potential for soil to be utilised for crop production. For Vertosols, a lack of stable aggregates >100 [micro]m in diameter may result in the formation of surface seals, affecting the ability of soil to receive and store water and to enable the rapid and uniform emergence of crops (Loch 1994). Of more concern is the complete disintegration of aggregates resulting in an undesirable massive structure (Field et al. 1997). Commonly, the assessment of aggregate fractions of various soil types requires the use of procedures similar to those developed for assessing aggregate stability in water. Zanini et al. (1998) note that most methods of measuring aggregate stability are based on the assumption that soil possesses a minimum state of aggregation when exposed to wetting. The assessment of water stability of aggregates often involves a combination of wetting procedures and subsequent mechanical agitation with various energies (Le Bissonais 1996). The combination of wetting procedures and mechanical agitation not only allows the determination of differing degrees of aggregation, but also allows the mechanisms of breakdown to be distinguished. The instantaneous wetting of aggregates with rapid immersion is often used to assess the effect of entrapped air, whereas the slow pre-wetting of aggregates before immersion and subsequent mechanical agitation is often employed to assess the mechanical stability of aggregates independent of the effect of entrapped air (Martinez-Mena et al. 1998).
These differences associated with various water stability procedures were exploited by Oades and Waters (1991) in their systematic study to assess and define the components of an aggregate hierarchy. The presence of smaller aggregates that are more stable than their larger counterparts defines the ordered system described by the aggregate hierarchy theory (Raine and So 1997). Although the size classes used to define components of aggregate hierarchies may be considered arbitrary and experiential (Tippkotter 1994), the contemporary concepts proposed have been widely accepted by researchers. Concerning the assessment of aggregation in Vertosols, Coughlan (1984) presented a detailed study of the dry and wet stability of aggregate fractions when considering the structure of Vertosols but there was no explicit statement defining a detailed aggregate hierarchy for these soils. The study of aggregation in Vertosols by Loch (1994) was based on the measure of a particular size fraction perceived to influence the potential for surface seal formation. Some Vertosols were included in the work of Chan and Mullins (1994) who studied the effect of antecedent moisture content and wetting rate on slaking characteristics of soil. They stated that the use of the term 'microaggregate' to describe slaked particles is based on the implicit assumption that the fine particles produced by slaking require more mechanical energy to disrupt than do larger aggregates. Raine and So (1997) noted that end-over-end shaking was unable to disrupt microaggregates 2-20 [micro]m in diameter, whereas ultrasonic agitation resulted in aggregates being fully dispersed. They stated that these results concur with Oades and Waters' (1991) definition of an aggregate hierarchy. Field and Minasny (1999) modelled the liberation and subsequent dispersion of 2-20 [micro]m diameter microaggregates from 6 Vertosol samples exposed to increasing energy supplied by ultrasonic agitation. The limited size fractions analysed within each study have not resulted in a comprehensive description of an aggregate hierarchy for the Vertosols. In a review presented by Dalal and Bridge (1996), the discussion of Alfisols and Mollisols (equivalent to Chromosols and Dermosols, respectively, in the Australian Soil Classification scheme of Isbell et al. 1997) is strongly linked to the contemporary definitions of the aggregate hierarchy, yet they did not demonstrate such a link when discussing aggregation for Vertosols. It is apparent that a systematic study, such as that presented by Oades and Waters (1991), is required for the Vertosols.
In addition to the size classes that comprise an aggregate hierarchy, much has been made of the morphological features of individual aggregate components of an aggregate hierarchy and their potential spatial relationship (Tisdall 1996). However, there are few reports in the literature of the effects that various aggregate liberation procedures may have on the resulting aggregate morphology observed, and this is especially true for the Vertosols.
Thus, this paper describes the efficacy of selected aggregate liberation methods to liberate aggregates from Vertosol samples and discusses the perceived degree of aggregate deformation and alteration produced by the aggregate liberation methods. The paper also aims to ascertain the existence of aggregate hierarchies in the Vertosols using predefined, pragmatic aggregate size fractions and these aggregate liberation methods.
Materials and methods
Samples from the top 50 mm of 3 Vertosols were collected for this study. The 3 samples used were collected from a dryland cotton site (QC, 150 [degrees] 57'E, 26 [degrees] 55'S) near Dalby (Queensland) and 2 irrigated cotton sites near Warren (WC, 147 [degrees] 46'E, 31 [degrees] 46'S) and Narrabri (NC, 149 [degrees] 32'E, 30 [degrees] 11'S) in New South Wales. Key physical and chemical properties of these topsoils are given in Table 1. Particle size analysis (PSA) was determined using the pipette method described by Gee and Bauder (1986). Exchangeable sodium percentage (ESP) was calculated from the measured exchangeable cations determined by a modification of the method described by Tucker and Beatty (1974). Total organic carbon was determined on solid-state samples by LECO high temperature carbon analyser as described by Merry and Spouncer (1988).
Half of the samples were air-dried and then gently ground and sieved to obtain a 1-2-mm-diameter fraction. The remaining half of each sample was broken gently by hand while still moist to obtain some 'non-ground aggregates'. These non-ground aggregates were then air-dried and stored separately for analysis. All ground and non-ground samples were stored in air-tight containers in a constant temperature room at 20 [degrees] C until required.
Preparation of minimally disturbed aggregates for morphological study (C1)
To ascertain whether deformation of liberated aggregates occurred from immersion wetting and/or from subsequent mechanical inputs, it was necessary to develop a protocol to simultaneously observe the internal and external morphology of the aggregates without the imposition of these forces. To achieve this, 20-50-mm-diameter aggregates were manually selected from the non-ground samples and either tension-wetted to -1 kPa or allowed to remain in an air-dry state. These aggregates were frozen rapidly by immersion into liquid nitrogen after being adhered to an aluminium stub by araldite. The affixed frozen aggregates were then fractured with a razor blade and placed into a freeze drier for 24 h to sublimate the water. The freeze-dried aggregates were then prepared for scanning electron microscopy (SEM).
Preparation of morphologically deformed aggregates for morphological study (C2)
This method involved immersion of a 5-mm-diameter aggregate of soil into a conical flask containing 30 mL of deionised water. The conical flask was placed on a shaking table and swirled for 24 h at a rate of 117 r.p.m. The contents of the conical flask were then treated with the standard aggregate washing method described below. This method is analogous to movement of material by flowing bodies of water, whereby the greater the distance travelled by the particles, the greater the degree of rounding and smoothing (Clarke and Cook 1986). Thus, it was expected that this method would result in notably rounded and smoothed aggregates which could be used as a standard to assess the potential aggregate deformation caused when the other aggregate liberation procedures described below were employed.
Standard aggregate washing method (SAWM)
This method was developed to minimise the coating of aggregates by dispersed clay upon drying. The suspension and sediments from a treatment were placed into a Buchner funnel containing 3-[micro]m Millipore paper and deionised water. The sediment and suspension were gently swirled around and subsequently, a vacuum was applied to remove the clay suspension. Deionised water was then added via a wash bottle down the internal side of the Buchner funnel to pond the sediment. Once again the sediment and suspension were swirled around followed by replication of the vacuum. This process was repeated a further 4 times. Once the washing was completed, the Millipore paper was removed and placed in a Petri dish housed in a desiccator to obtain equilibrium at 54.4% relative humidity. The Millipore paper was then gently tapped so that some aggregates rolled onto SEM stubs coated with double-sided sticky tape. An additional stub was also prepared by removing a small area of the Millipore paper and adhering to an SEM stub.
Aggregate liberation methods
Aggregate slaking in water (ASW)
This method is a modification of the method developed by Singer et al. (1994) to slake the aggregates and disperse the clay. Deionised water was ponded to a depth of 20 mm on top of a 3-[micro]m Millipore paper in a Buchner funnel. A 5-mm-diameter air-dry aggregate of soil was placed into the deionised water and left for 2 h and subsequently treated according to the SAWM.
Modified wet sieving (MWS)
This method is based on that of Cambardella and Elliot (1993). Air-dry or tension-wetted (-1 kPa) ground (1-2 mm) soil samples were passed sequentially through a series of 3 sieves to obtain the > 250, 100-250, 50-100, and < 50 [micro]m fractions (MW[S.sub.ad] and MW[S.sub.w], respectively). This was achieved by suspending the sample on the largest sieve immersed in deionised water for 5 min before sieving. Aggregate separation was accomplished by moving the sieve 30 mm vertically 50 times within 2 min. Material that passed through the sieve was poured onto the next finer sieve and the process repeated. Material remaining on the sieve was gently washed into a container with deionised water and stored for preparation using the SAWM. The < 50 [micro]m fraction was transferred to a 600-mL glass beaker and left for the appropriate time for the > 2 [micro]m particles to settle to a depth of 100 mm. The suspension was siphoned off using an up-end siphon attached to a vacuum, and stored. A subsample of the sediment was then treated using the SAWM.
End-over-end shaking (EOE)
Using the recommendations given after the assessment of end-over-end shaking methods by So et al. (1997), the following method was adopted. In this method, 6 g of the ground (1-2 mm) air-dry soil was immersed in 100 mL of deionised water. The samples were then subjected to end-over-end agitation in 250-mL centrifuge bottles with a maximum internal diameter of 55 mm, a height of 120 mm, and an air gap of 80 mm. Shaking was conducted at 30 r.p.m. for 30 min. After shaking, the sediment and suspension were transferred to a 600-mL glass beaker and left for the appropriate time for the > 2 [micro]m particles to settle to a depth of 100 mm. The suspension was siphoned off using an up-ended siphon and vacuum, and stored. A subsample of the sediment was then treated using the SAWM.
Ultrasonic agitation (UA)
Samples for ultrasonic agitation were prepared by immersion of the ground (1-2 mm) air-dry equivalent of 4 g oven-dry soil (105[degrees]C) into 20 mL of deionised water. Specifications for the sample container, power output, and ultrasonic instrumentation were the same as those presented in Field and Minasny (1999). Each suspension was treated with ultrasound for 2 periods of 60 or 120 s. After sonification, the sediment and suspension were transferred to a 600-mL glass beaker and left for the appropriate time for the > 2 [micro]m particles to settle to a depth of 100 mm. The suspension was siphoned off using an up-ended siphon and vacuum, and stored. A subsample of the sediment was then treated using the SAWM.
Sample coating and scanning electron microscopy conditions
Each of the stubs prepared from the SAWM was coated with a 20-nm-thick coating of platinum using a 45[degrees] rotation. These were then stored air-dry in a desiccator until needed. SEM was carried out on a Philips XL30 with integrated EDS (DX4).
Quantification of aggregate separates
Additional samples were treated using the methods to quantify the degree of aggregate liberation and dispersion. The small amount of material produced by the aggregate slaking in water (ASW) method meant that quantification was not feasible. Duplicate samples treated using the modified wet-sieving air-dry (MW[S.sub.ad]) and tension-wetted (MW[S.sub.w], end-over-end shaking (EOE), and ultrasonic agitation (UA) methods were followed by manual wet-sieving and sedimentation to obtain the following size fractions: > 250, 100 250, 50 100, 20 50, 2 20, < 2, and < 0.2 [micro]m. PSA was determined using the pipette method described by Gee and Bauder (1986) and partitioned using the same size fractions described above.
Results and discussion
Efficacy of the aggregate liberating methods to liberate various aggregate fractions
The particle-size distributions for the MWS, EOE, and UA methods, and for totally dispersed samples, are presented in Fig. 1. Considering that the ASW method involved the simple immersion of air-dry aggregates in water, it was assumed that this procedure would produce proportions of larger aggregates equal to or greater than the MW[S.sub.ad] method. Observation of the aggregates produced by the ASW using SEM indicated that this was the case.
[FIGURE 1 OMITTED]
For all samples, at least 50% of the soil was found to consist of aggregates > 250 [micro]m in diameter following the MW[S.sub.w] treatment. By comparison, there was a relative reduction in the soil material present in the > 250 [micro]m fraction following the MW[S.sub.ad] treatment, and a concomitant increase in soil material residing in the 100-250 [micro]m fraction. The pre-wetting of the samples before short periods of wet-sieving was performed as a best estimate to elucidate mechanisms associated with aggregate breakdown independent of mechanical abrasion. For example, air occlusion and swelling phenomena are both known to increase with wetting (Zanini et al. 1998). Thus, it may be assumed that the tension-wetting of the aggregates reduced the effects of occluded air, reducing the degree of disruption for larger aggregates. The wetting rate also affects aggregate disruption through differential swelling. The mechanisms of differential swelling may be responsible for aggregate liberation considering the significant shrink-swell character associated with Vertosols (Chan and Mullins 1994). Although the rate of wetting cannot be controlled by tension-wetting (Loch 1994), it appears that the wetting rate associated with the MW[S.sub.w] method was sufficiently small to minimise any possible disruption of the larger fractions that may be caused by differential swelling.
In addition to the energies associated with wetting, aggregate disruption and dispersion are also affected by increases in mechanical energies to which the soil is exposed. Considering that the same wetting procedures were used for the MW[S.sub.ad] and EOE methods, comparison of these 2 procedures is appropriate to determine the effect of additional energy associated with mechanical abrasion. The EOE method resulted in a reduction of soil material in the > 100 [micro]m fractions and an accumulation of soil material residing in the < 100 [micro]m fraction compared to the MW[S.sub.ad]. However, the mass of soil material > 100 [micro]m in diameter was still far in excess of that measured by the PSA for each of the topsoils (Fig. 1), indicating that the mechanical energy supplied by the EOE method was not sufficient to disrupt all the aggregated soil material. The mechanical abrasion associated with the EOE method seems to have increased the amount of the 2-50 [micro]m fraction compared with the MW[S.sub.ad] method.
For all samples exposed to the agitation supplied by the 2 time periods used in the UA method, there is a significant decrease in the mass of material > 50 [micro]m in diameter compared with that of the other methods. With the exception of the Narrabri sample, it can be observed in Fig. 1 that energy supplied by the UA method resulted in a concentration of aggregated material in the 2-20 [micro]m fraction. The UA method has also resulted in an observable increase in the mass of the < 2 [micro]m fraction compared with that of the MWS and EOE procedures.
It appears that the use of the ASW and MWS methods would be appropriate for the observation of aggregated material > 100 [micro]m, whereas the EOE method is more appropriate for aggregates within the 2-100 [micro]m fraction. For the isolation of the aggregated soil material < 50 [micro]m in diameter, short periods of ultrasonic agitation appear most suitable. To verify the utility of these aggregate liberation methods, any morphological deformation of the liberated aggregate fractions should be determined.
Description of the minimally disrupted (C1) and morphologically deformed (C2) aggregates', and evaluation of the SAWM method, using SEM
Examples of aggregates produced by the C1 method are shown in Fig. 2. Examination of the freeze/dry fracture faces of the aggregates exhibited several distinct morphologies regarding the orientation of clay. It can be seen in Fig. 2a that a pore, labelled 'p', dominates the lower half of the micrograph with a smooth undulating surface. Such surfaces can be indicative of clay coatings, as defined by Sullivan and Koppi (1994). Fine clay-coated roots, labelled 'cr', are also observable in the pore space. The upper half of the micrograph shows a fracture surface revealing small clusters of discrete aggregates. The label 'ag' points out several of these discrete aggregates. Generally, these aggregates are 10-50 [micro]m in diameter, exhibiting a moderately spheroidal morphology, and the surfaces are sub-angular to sub-rounded. The micrograph presented in Fig. 2b shows the morphology of the 10-50 [micro]m aggregates. The clustering of the clay gives a rugose appearance, i.e. a wrinkled or corrugated appearance to the 10-50 [micro]m aggregates. The morphology of the clay within these microaggregates is a characteristic associated with the clustering of quasi-crystals of smectite described by Oades and Waters (1991). The freeze-drying of the aggregates results in a temporary hardening of the aggregates, making them brittle rather than plastic. The brittle nature of the aggregates would probably have prevented any aggregate compression when exposed to the force of fracturing. If compression had occurred it is expected that the aggregates would have appeared platy, which was not observed (Fig. 2b). There also was no evidence of scoring of the aggregate surfaces by the passing of the blade used for fracturing. It was therefore assumed that a brittle fracture had occurred with little or no deformation of the microaggregates.
[FIGURE 2A-2B OMITTED]
The aggregates liberated by the C2 method are shown in Fig. 3. In Fig. 3a, 3 aggregates from Narrabri are presented that are quite rounded with smooth surfaces. Various entrances to pores and pits, labelled 'ep', which are possibly the result of dislodged mineral grains, are observable. The aggregates have no coatings of clay on their surfaces, and thus the smooth appearance may be the result of the surface being polished by the abrading action associated with aggregate collisions. An example aggregate from Warren produced by the C2 method is presented in Fig. 3b. Here, it is difficult to determine if there has been any major degree of rounding, as there is the presence of coating clay (labelled csc). Similar aggregate morphologies were observed to varying degrees for all samples subjected to the C2 method. For aggregates < 100 [micro]m in diameter the degree of rounding was variable. The increased probability of larger aggregates tending to roll around the bottom of the beaker may have resulted in an increased rounding of these aggregates. The morphology of selected aggregates isolated by this method contrasted with those prepared by the C1 method. It appears that the C2 method produced an altered morphology for selected aggregates that may be used as a standard to assess the degree of morphological deformation of aggregates produced by the 4 methods under investigation.
[FIGURES 3A-3B OMITTED]
Liberated aggregates that were prepared when a subsample of the suspension was placed on a stub and allowed to dry are shown in Fig. 3c. It is apparent that some of the dispersed clay in the suspension coated the aggregate on drying, forming a continuous film between the aggregate and stub (f). This coating was still occasionally observed for samples where clay had been siphoned off in an attempt to remove this dispersed material. However, if we consider the subsequent micrographs displaying the aggregates produced by the various liberation techniques, and pre-treated using the SAWM, no such coating onto the stubs is observable (Fig. 4). Also, the pores and fibres that form the Millipore paper shown in some of the micrographs are still distinguishable. Based on these observations it appears that the SAWM is largely successful in preventing coating of the aggregates with suspended clay upon drying.
[FIGURES 3C-4 OMITTED]
Observation of aggregates produced by the four aggregate liberating methods
Selected examples of aggregates from Narrabri, Warren, and Dalby produced by the ASW, MWS, EOE, and UA methods, combined with the SAWM, are presented in Fig. 4. These micrographs demonstrate that some of the aggregate liberating methods may have resulted in morphological deformation of aggregates, whereas others produced apparently unaltered aggregates. Figure 4a is an example of an aggregate > 250 [micro]m in diameter, composed of smaller aggregates < 100 [micro]m in diameter labelled 'ma', from the Warren site and produced by the ASW method. The smaller aggregates exhibit a spheroidal morphology with sub-angular to sub-rounded surfaces. This shape is akin to those aggregates produced by the C1 method. Similar morphologies were also observed for the aggregates from Narrabri and Dalby produced by the ASW method.
A cluster of aggregates produced by the MWS method from the Narrabri sample is presented in Fig. 4b. The 2 aggregates labelled '[ag.sub.1]' and '[ag.sub.2]' do not appear to exhibit any evidence of rounding, whereas it is not clear whether the aggregate labelled '[ag.sub.3]' has experienced rounding. The possible deformation of the Narrabri samples was observed occasionally for other replicates and may have been caused by abrasion of the aggregates as they passed through the mesh of the sieves. Aggregates produced by the MWS method for the Warren and Dalby samples consistently displayed morphologies similar to those produced by the C1 method and thus there is little evidence of morphological deformation.
[FIGURE 4B OMITTED]
In contrast, when aggregates prepared by the EOE method were examined, the morphology observed depended on the size range studied. Larger aggregates, including those > 50 [micro]m in diameter, were rounded and there was some evidence of smoothing. This feature is expressed by the aggregate from Dalby presented in Fig. 4c. The morphology is very similar to that of the selected aggregates from the C2 method. An example of an aggregate < 50 [micro]m in diameter from Narrabri is presented in Fig. 4d, and is shown to be rugose, with little sign of rounding or smoothing of the aggregates. Simplistically, it may be assumed that the degree of rounding and smoothing of aggregate surfaces would depend on (i) the abrading ability of the water and collision intensity between individual particles, and (ii) the time for which the surface of an aggregate is exposed to the abrasion of water flow or particle collision (So et al. 1997). It seems possible that the intensity of the energy supplied by the EOE method was not able to disrupt these large aggregates, but the rolling and falling with each inversion of the cylinders may have resulted in the gradual rounding and smoothing of the aggregates. The reason for the lack of morphological deformation of the smaller aggregates is less clear. It may be the case that these aggregates were not liberated immediately and therefore were not exposed to the forces associated with the abrasion of water flow or particle collision, limiting the degree of deformation. Considering that the intensity of the collision between particles would be dependent on the mass and velocity of these particles, as the EOE method proceeds and large aggregates are broken up, the intensity of collisions may decrease accordingly.
[FIGURES 4C-4D OMITTED]
The UA method resulted in aggregates > 100 [micro]m diameter being disrupted, so observation of aggregates larger than this size was impossible (Fig. 1). An example of a Narrabri aggregate liberated by the UA (60 s) method is presented in Fig. 4e. The morphology of the aggregate is characteristic of the clustering of quasi-crystals of clay. A micrograph showing an aggregate from Dalby produced by the UA (120 s) method is presented in Fig. 4f The rugose appearance characteristic of the clay microstructures produced by the quasi-crystals of clay are labelled 'qc'. The presence of discrete mineral grains, characterised by a smoother appearance, is identified by the label 'mg'. It is apparent that the morphologies of the aggregates show little evidence of rounding or smoothing. Similar morphologies are observed for the Warren samples subjected to the UA method.
[FIGURES 4E-4F OMITTED]
From the observations presented for the 4 liberation methods it is not possible to develop a general statement concerning the ability of the MWS method to alter the morphology of the samples, as only aggregates from Narrabri seem to have been be affected. The EOE method, in particular, show evidence of morphological deformation for all samples, but this appeared to be applicable only to aggregates > 50 [micro]m in diameter. The ASW and UA methods showed little evidence of morphological deformation. Thus, depending on the degree of liberation required, the ASW and UA methods, and possibly the MWS method, seem to be the most appropriate methods for producing unaltered aggregates for morphological interpretation by use of SEM.
Some comments on the aggregate hierarchy of the Vertosols studied
Keeping in mind that the size classes of aggregate hierarchies may be considered arbitrary (Tippkotter 1994), the choice of the size fractions used to assess the efficacy of the methods to liberate Vertosol aggregates was influenced by the commonly accepted aggregate hierarchy subdivisions and soil management implications. The choices of the cut-off at 250 [micro]m diameter and the 2-20 [micro]m subdivision are in accordance with the traditional division of macroaggregates and microaggregates defined by Tisdall and Oades (1982) and Oades and Waters (1991). The choice of the cut-off at 100 [micro]m diameter is influenced by soil management implications as presented by Loch (1994), who suggested that an increase in aggregate breakdown to < 100 [micro]m diameter increases the potential of surface crust formation, which will impede water infiltration. The subdivision at < 0.2 [micro]m diameter is based on the assumption that the < 2 [micro]m fractions of Vertosols consist of clusters of fine clay that would require greater energy for dispersion than is required to disperse the conventionally defined clay fraction (Coulombe et al. 1996).
The use of the MW[S.sub.w] method resulted in the liberation of aggregates generally > 250 [micro]m in diameter, whereas the MW[S.sub.ad] method resulted in the liberation of aggregates < 250 [micro]m in diameter (Fig. 1), which is similar to results presented by Oades and Waters (1991) for several Chromosols and Dermosols. As previously discussed, Fig. 4a illustrates that, for the Vertosols, macroaggregates > 250 [micro]m in diameter are composed of microaggregates < 100 [micro]m in diameter. The next stepwise breakdown of the aggregates was produced by the EOE method, with a large proportion of the aggregates accumulating in the fraction < 100 [micro]m diameter (Fig. 1). The use of the UA method resulted in most of the stable aggregates residing in the 2 20 [micro]m fraction (Fig. 1). The UA (120 s) method did not disperse all the clay and dispersed very little of the fine clay (< 0.2 [micro]m) compared with the PSA method. It is not clear from the data (Fig. 1) whether the disruption of 2-20 [micro]m microaggregates results in the proportional increase of both the coarse and fine clay concentrations. Alternatively, the disruption of these aggregates may result in the accumulation of aggregations in the coarse clay fraction that require an incremental increase in energy to cause dispersion.
From the data presented, the aggregate hierarchy for the Vertosols may be divided into fractions > 250, 100-250, 20-100, 2-20, and < 2 [micro]m in diameter. The notion that all soil aggregates must break down via predetermined size fractions may not be precise, therefore ignoring information regarding the alternate distributions of aggregation if different size classes were chosen. As previously outlined, the size classes chosen were governed by the commonly accepted aggregate hierarchy subdivisions and by soil management implications. The use of narrower size class ranges may have identified other intermediate hierarchy steps for the Vertosols studied.
The incremental increases in energy supplied by each of the aggregate liberating methods resulted in the liberation of progressively finer aggregate fractions. The data suggest that for Vertosols, the ASW and MWS methods are suitable for the liberation of aggregates > 100 [micro]m in diameter for morphological description. Evidence was presented to suggest that the EOE method might cause morphological deformation of the samples, depending on their size, and thus was not recommended for preparation of samples for micromorphological study. The more intense energy applied by the UA method was appropriate for the liberation of aggregates < 50 [micro]m in diameter.
Information from the various liberation methods was used to characterise the aggregate components that constitute a proposed aggregate hierarchy for the Vertosols. The stepwise breakdown of aggregates with the assumed increases in energy applied by each of the liberation methods satisfies the accepted definition of an aggregate hierarchy. Constrained by the choice of liberation methods, it is proposed that the aggregate hierarchy for the Vertosols studied may be divided into fractions > 250, 100-250, 20-100, and 2-20 [micro]m in diameter. It is not clear from the data presented here whether the < 2 [micro]m fraction consists of an aggregate component 0.2-2 [micro]m in diameter.
Table 1. A summary of important physical and chemical properties for each of the three Vertosol topsoils (0-50 mm) sampled from Narrabri (NC), Warren (WC), and Dalby (QC) Particle-size distribution (dag/kg) Sample >200 20-200 2-20 0.2-2 <0.2 [micro]m [micro]m [micro]m [micro]m [micro]m NC 5 8 17 26 44 WC 8 25 14 25 28 QC 15 27 16 9 33 Sample ESP Total organic carbon (dag/kg) NC 10.7 0.70 WC 1.4 0.60 QC 0.8 0.99
The authors gratefully acknowledge the preliminary studies conducted by C. E. Wood, which proved useful when developing the standard aggregate washing method, and the journal referees who gave several useful suggestions. Funding to support the senior author was provided by the Australian Cotton Co-operative Research Centre.
Cambardella CA, Elliot ET (1993) Methods for physical separation and characterization of soil organic matter fractions. Geoderma 56, 449-457. doi: 10.1016/0016-7061(93)90126-6
Chan KY, Mullins CE (1994) Slaking characteristics of some Australian and British soils. European Journal of Soil Science 45, 273-283.
Clarke IF, Cook BJ (1986) 'Perspectives of the Earth.' (Australian Academy of Sciences: Canberra)
Coughlan KJ (1984) The structure of Vertisols. In 'The properties and utilisation of cracking clay soils'. (Eds JW McGarity, EH Holt, HB So) pp. 87-96. (University of New England: Armidale, NSW)
Coulombe CE, Dixon JB, Wilding LP (1996) Mineralogy and chemistry of Vertisols. In 'Vertisols and technologies for their management'. (Eds N Ahmad, A Mermut) pp. 115-200. (Elsevier Science)
Dalal RC, Bridge BJ (1996) Aggregation and organic matter storage in sub-humid and semi-arid soils. In 'Structure and organic matter storage in agricultural soils'. (Eds CM Carter, BA Stewart) pp. 263-307. (CRC Press Inc.: Boca Raton, FL)
Field DJ, McKenzie DC, Koppi AJ (1997) Development of an improved Vertisol stability test for SOILpak. Australian Journal of Soil Research 35, 843-852.
Field DJ, Minasny B (1999) A description of aggregate liberation and dispersion in A horizons of Australian Vertisols by ultrasonic agitation. Geoderma 91, 11-26. doi:10.1016/S0016-7061(98) 00142-6
Gee GW, Bauder JW (1986) Particle size analysis. In 'Methods of soil analysis. Part 1'. (Ed. A Klute) (American Society of Agronomy, Soil Science Society of America: Madison, WI)
Isbell RF, McDonald WS, Ashton LJ (1997) Concepts and Rationale of the Australian Soil Classification (ACLEP/CSIRO Land and Water: Canberra)
Le Bissonais Y (1996) Aggregate stability and assessment of soil crustability and erodibility: I. Theory and methodology. European Journal of Soil Science 47, 425-437.
Loch RJ (1994) A method for measuring aggregate water stability with relevance to surface seal development. Australian Journal of Soil Research 32, 687-700.
Martinez-Mena M, Williams AG, Ternan JL, Fitzjohn C (1998) Role of antecedent water content on aggregate stability in a semi-arid environment. Soil and Tillage Research 48, 71-80. doi:10. 1016/S0167-1987(98)00131-7
Merry RH, Spouncer LR (1988) The measurement of carbon in soils using a microprocessor-controlled resistance furnace. Communications in Soil Science and Plant Science 19, 702-720.
Oades JM, Waters AG (1991) Aggregate hierarchy in soil. Australian Journal of Soil Research 29, 815-828.
Raine SR, So HB (1997) An investigation of the relationship between dispersion power, and mechanical energy using the end-over-end shaking and ultrasonic methods of aggregate stability assessment. Australian Journal of Soil Research 35, 41-53.
Singer A, Kirsten WFA, Buhmann C (1994) Clay dispersivity and crusting of soils determined by Buchner funnel extraction. Australian Journal of Soil Research 32, 465-470.
So HB, Cook GD, Raine SR (1997) An examination of the end-over-end shaking techniques for measuring soil dispersion. Australian Journal of Soil Research 35, 31-39.
Sullivan LA, Koppi AJ (1994) Submicromorphology of undulating lustrous-faced peds from some clayey soils. Soil Science 158, 218-223.
Tippkotter R (1994) The effects of ultrasound on the stability of mesoaggregates (60-2000 [micro]m). Zeitschrift Pflanzenernahr Bodenkunde 157, 99-104.
Tisdall JM (1996) Formation of soil aggregates and accumulation of soil organic matter. In 'Structure and organic matter storage in agricultural soils'. (Eds CM Carter, BA Stewart) pp. 57-96. (CRC Press Inc.: Boca Raton, FL)
Tisdall JM, Oades JM (1982) Organic matter and water stable aggregates in soil. Journal of Soil Science 33, 141-163.
Tucker BM, Beatty HJ (1974) Exchangeable cations and cation exchange capacity. In "Methods for analysis of irrigated soils'. (Ed. J Loveday) Technical Communication No. 54, Commonwealth Bureau of Soils.
Zanini E, Bonifacio E, Albertson JD, Nielson DR (1998) Topsoil aggregate breakdown under water-saturated conditions. Soil Science 163, 288-299. doi: 10.1097/00010694-199804000-00004
Damien J. Field (A,E), Leigh A. Sullivan (B), Stephen R. Cattle (C), and Anthony J. Koppi (D)
(A) Australian Cotton Co-operative Research Centre, The University of Sydney, NSW 2006, Australia.
(B) School of Environmental Science and Management, Southern Cross University, Lismore, NSW 2840, Australia.
(C) School of Land, Water and Crop Sciences, The University of Sydney, NSW 2006, Australia.
(D) Educational and Development Technology Centre, The University of New South Wales, NSW 2052, Australia.
(E) Corresponding author; email: email@example.com
Manuscript received 7 May 2003, accepted 18 September 2003
|Printer friendly Cite/link Email Feedback|
|Author:||Field, Damien J.; Sullivan, Leigh A.; Cattle, Stephen R.; Koppi, Anthony J.|
|Publication:||Australian Journal of Soil Research|
|Date:||Jan 1, 2004|
|Previous Article:||Herbicide loss in runoff: effects of herbicide properties, slope, and rainfall intensity.|
|Next Article:||The effects of clay and sand additions on the strength of sandy topsoils.|