Printer Friendly

Do sand dunes of the lower Lachlan floodplain contain the same dust that produced parna?

Introduction

The deposition of aeolian dust is known to have been an active factor in the process of soil formation and landscape evolution on every continent. AEolian dust has been broadly defined as a suspension of wind-blown particles of diameter less than 100 [micro]m, or a deposit of such particles (Pye 1987). Worldwide, the most common manifestation of aeolian dust deposition is the formation of loess profiles. Whilst Australia lacks thick loess mantles found elsewhere, the extensive dune fields of Australia's arid interior attest to the importance of aeolian processes in the evolution of the Australian landscape (Wasson et al. 1988). Significant quantities of dust must also have been entrained during storms associated with the formation of these dune fields, and subsequently carried far beyond the desert margin. Virtually all Australian soils have a component of dust accession, although its significance at a given location can be difficult to determine (McKenzie et al. 2004) and as such the true extent and nature of aeolian dust deposits in Australia are not entirely identified and characterised (Hesse and McTanish 2003).

The dominant model of aeolian dust deposits in southeastern Australia is the clay-rich, calcareous 'parna', as pioneered by Butler (1956). He proposed that parna occurs across southern NSW and northern Victoria as a relatively uniform, thin sheet covering the whole landscape irrespective of relief, and that it is distinct from 'loess' in both its composition and origin. Butler and Hutton (1956) argued that this parna sheet was formed by the erosion of soils in the dune zone of western NSW and Victoria, and the subsequent aeolian transport and deposition of clayey aggregates and companion grains downwind. Implicit in this model is the belief that the parna-forming dust was composed chiefly of secondary minerals (30-70% clay, some calcium carbonate), in contrast to the often glacially derived primary mineral fragments that form characteristic loess. Accordingly, it has been asserted that soil profiles formed from parna differ from those formed from glacially derived loess because they have effectively already passed through one cycle of weathering, being composed almost entirely of secondary minerals.

AEolian dust has been identified and described in a number of soils throughout southern Australia, but not all of these aeolian dust deposits conform to the benchmark model of parna. Considerable variation in morphological and mineralogical properties of dust has been reported, and this variation is largely related to the nature of, and distance from, source areas. Fine-grained, quartz-rich aeolian dust additions to soils of eastern and northern New South Wales (Cattle et al. 2002; Hesse and McTanish 2003; Hesse et al. 2003) differ considerably from the clay-rich, calcareous parna of southern NSW as described by Butler (1956), Beattie (1970), and Chen (2001), amongst others. As a consequence, there has been some debate as to the validity of the parna model, with some authors (e.g. Hesse and McTanish 2003) suggesting that there is little clear evidence of pelletal clay aggregates being transported long distances as dust. Dare-Edwards (1984) and Hesse and McTanish (2003) suggest that aeolian dust deposits of Australia bear a strong similarity to conventional definitions of desert-derived loess, and contend that parna should be considered as 'desert loess' or 'loessic clay'.

On the north-western fringe of the proposed zone of parna deposition (Butler 1956), the presence of quartz-rich sand dunes offers an ideal opportunity to investigate whether pelletal dust exists in a relatively inert medium. These sand dunes, which border palaeochannel tributaries of the lower Lachlan River, have been observed to contain an apparently clay-enriched zone in the upper part, and a zone of carbonate accumulation at the base. Such features, repeated across a number of partially deflated dunes, suggest a significant aeolian dust component. In a study of dunes showing similar features at Wagga Wagga in southern New South Wales, Chen et al. (2002) suggested that the dunes had functioned as dust traps, and hence possibly contained the material that constitutes parna. If the fine material in sand dunes of the lower Lachlan is the same type of dust that produced parna, then these dunes provide a unique opportunity to characterise relatively pure deposits of such material.

Thus, the aims of this study were to: (i) characterise the morphological and mineralogical properties of fine material trapped in 3 source-bordering sand dunes on the Lachlan floodplain at Hillston; (ii) determine whether this fine material is of a local or allochthonous origin; and (iii) ascertain whether the dunes contain intact pelletal clay aggregates consistent with Butler's (1956) model of parna formation.

Materials and methods

Regional setting and study sites

Three source-bordering sand dunes in the lower Lachlan Valley of south-western New South Wales were selected as sampling sites for this study. Two were located on a property 10km north of the township of Hillston and the third on a property 20km west of the town. These sampling sites will be referred to as 'Embagga' (6 306 388 m N, 0 360 009 m E), 'Kongong' (6 307 546 m N, 0 360 931 m E), and 'Florabel' (6 295 644 m N, 0 339 686 m E).

The climate of the study region is best described as semi-arid and temperate, with an evenly distributed average annual rainfall of 367 mm, and average potential evaporation of 1825 mm. At Hillston, average summer daily temperatures range from 18.1 to 33.0[degrees]C and average winter daily temperatures range from 3.9 to 14.8[degrees]C (Bureau of Meteorology 1988).

The sampling region is situated within the partly active alluvial floodplain of the lower Lachlan River catchment, and the terrain is flat to slightly undulating. The alluvial plains are dominated by Vertosols (Isbell 1996), although source-bordering dunes (sand-mounds), often up to 10 m high, are a repeated feature within this landscape. These dunes were formed by the aeolian reworking of palaeochannel bed-load sand (Cattle et al. 2003). A common feature of these sand dunes is a deflated structure, exposing the presence of a thick, reddish, more clayey layer in the upper part and an accumulation of carbonates at the base of the dune. Three dunes exhibiting these distinct features were chosen as the sampling sites for this work (Embagga, Kongong, Florabel).

Sampling

At each sampling site, the macromorphological features of the dunes were described, including the 3 distinct phases (summarised in Fig. 1) that were consistent for all dunes. As indicated in Table 1, the uppermost phase of each dune (phase 's') consisted of coarse, mostly loose sand showing minimal structural development, except at the Embagga site, where a poorly developed, very weak and brittle, platy structure was evident. The second phase (denoted as 'cs') comprised clay-enriched sand and showed some variation in structure, with the Embagga and Florabel sites being apedal massive in this phase, and the Kongong site exhibiting very weak, platy peds. In the Florabel dune only, some distinct, thin, brightly coloured, clayey lamellae (phase 'l') were present at the base of the cs phase, and were in turn underlain by a further phase of clay and silt-enriched sand (phase 'sl'). At the base of each dune, a phase comprising coarse, calcareous sand (phase 'k'), including accumulations of carbonate structures classed as 'glaebules', was present.

[FIGURE 1 OMITTED]

Granulometric analyses

The particle size distribution of soil material from each phase in each of the 3 dunes was determined using the pipette method (Gee and Bauder 1986) for material <2 [micro]m, and wet sieving for material > 75 [micro]m. The quantified particle size fractions were: <2 [micro]m diameter ('clay'), 2-75 [micro]m (described here as 'silt'), 75-200 [micro]m ('fine sand'), and 200-2000 [micro]m ('coarse sand').

In addition, the non-clay fraction (2-400 [micro]m) of 4 samples was analysed using a Coulter Multisizer 3, producing a high-resolution particle size distribution encompassing 256 size classes per sample (McTainsh et al. 1997). The 4 samples analysed were taken from the cs phase of each of the 3 dunes and from the l phase of the Florabel dune. For each sample, analysis was performed on both soil that had been fully dispersed in Calgon, and on soil that had only been minimally dispersed by wet sieving.

Clay mineral suite

Clay suspensions obtained during the granulometric analyses were retained and used to produce oriented aggregates for X-ray diffraction analysis. For each soil sample, the clay suspension was applied to 2 separate ceramic tiles under suction. A 2 mol/L solution of KCl was then applied to one tile of each pair in 2 washing treatments, and a 2 mol/L solution of Mg[Cl.sub.2] applied to the other tile in the same way. All tiles were then washed 4 times with distilled water to remove excess salts (Brindley and Brown 1980).

The clay mineral suite of each sample was then determined by X-ray diffraction of the 2 tiles, using a Siemens Diffraktometer 5000. The samples washed with KCl were analysed air-dried and following heat treatments of 100, 300, and 550[degrees]C. Those treated with Mg[Cl.sub.2] were analysed air-dried and following solvation with glycerol. All analyses were conducted using Cu-K[alpha] radiation. KCl tile readings spanned the range 2-15[degrees] 2[theta], and Mg[Cl.sub.2] tile readings spanned the range 2-30[degrees] 20, with a step width of 0.02[degrees]. Clay mineral suite was determined for each phase of each of the 3 dunes.

Micromorphological analysis

Thin sections of undisturbed soil from the s, cs, and k phases of each dune, and the l phase of the Florabel dune, were prepared using soil collected in Kubiena tins. The soil was dried in the Kubiena tins (160 by 80 by 50mm) at 40[degrees]C and then impregnated with a polyester resin for 6 weeks. Vertically oriented thin sections 30-[micro]m thick were produced from the impregnated soil and mounted on glass slides. Optical properties of the thin sections were viewed under both plane-polarised light and cross-polarised light using a petrographic microscope, and photographed. The micromorphological descriptions follow Bullock et al. (1985).

Grain morphology

Soil samples were viewed by scanning electron microscopy (SEM). Randomly sampled loose material from the cs phase of Kongong and Florabel, the s phase of Kongong, and the l phase of Florabel was mounted on aluminium SEM stubs and observed and photographed using a Philips SEM 505.

Results and discussion

Granulometric properties

Particle size characteristics of the 3 common phases within the 3 sampling dunes are summarised in Fig. 2. All samples from the s, cs, and k phases of each dune show a similar pattern in particle size distribution, with sand comprising >75% of all particles, and coarse sand predominating over fine sand. The proportion of coarse and fine sand shows some variation, however. In general, there is an accumulation of fine sand in the cs phase, as indicated by smaller coarse sand: fine sand ratios. The Kongong dune contains a much larger proportion of fine sand (>20% in all phases) than the other 2 dunes. There is also a very large increase in silt content from the l phase of the Florabel dune to the underlying sl phase. In all dunes the clay content increases from the s to the cs phase, and decreases again in the k phase (Fig. 2), but the clay contents of the cs phases range only from 5 to 8%. The l phase of the Horabel dune contains the largest amount of clay (14%) of any phase in any dune.

[FIGURE 2 OMITTED]

Most stabilised aeolian sand dunes are known to contain a component of 'fines', or silt and clay-sized particles. The presence of these fines is related to a suite of post-depositional weathering processes, which includes the infiltration of aeolian dust (Tchakerian 1999). This process is largely accomplished by grain translocation, occurring with the movement of rainwater, dew, and groundwater, and mineral disaggregation over time (Tchakerian 1999). The infiltration of aeolian dust is thus a plausible explanation for the similar accumulation of fine material (silt and clay) in a specific zone of the 3 spatially separate dunes.

The high-resolution particle size distributions of samples from the cs phase of each dune and the l phase of the Florabel dune are shown in Fig. 3. For all samples, the population of grains in the 200-400 [micro]m range represents the coarse sand matrix of the dune. Whilst there are differences between the histograms produced at each site, there are consistent features confirming an aeolian dust component. All dunes contain a distinct mode in the silt size-range (20-60 [micro]m) and, for the Kongong and Florabel dune cs phases, these modes become more pronounced following dispersion (Fig. 3c-f).

[FIGURE 3 OMITTED]

Previous aeolian dust research in eastern Australia has interpreted the presence of a particle-size mode in the 30-60 [micro]m range as typical and indicative of aeolian dust accessions (e.g. Beattie 1970; Blackburn 1981; Walker et al. 1988; Cattle et al. 2002; Hesse et al. 2003). The shift in modal diameters following full dispersion of the Florabel cs phase soil (from 60 to 40 [micro]m, Fig. 3e and f) suggests that this material consists of 40-[micro]m quartz grains aggregated or coated with some clay and fine silt. Only after vigorous dispersion do the fines dissociate from the quartz grains (Fig. 3f). For the Kongong cs phase, the shift in modal diameter following dispersion suggests that the original sample consists of 50-60 [micro]m quartz grains aggregated or coated with substantial amounts of fine silt and clay (Fig. 3c and d). This subplastic character has been recognised in most parna deposits (Dare-Edwards 1984).

The Florabel l phase exhibits a relatively constant particle size distribution following both minimal and full dispersion (Fig. 3g and h), although there is a small amount of clay liberated from the silt mode by dispersion. The final nonclay particle size distribution of the Florabel l phase (Fig. 3h) is comparable with that described by Walker et al. (1988) in loessic subsoils of south-eastern Australia; a bimodal distribution with a main population of fine to coarse sand grains, and a smaller but pronounced population of grains in the coarse silt range.

The Embagga cs particle size histograms (Fig. 3a and b) are somewhat different from those of all other samples, due to the largely unimodal character, and also the dearth of a significant modal shift following full dispersion. There is only a minor mode in the Embagga coarse silt size range (40-60 [micro]m) (Fig. 3a and b). Erosion may provide an explanation for the lack of silt and clay, as the cs phase at Embagga is considerably more exposed and disturbed than the equivalent sites at Kongong and Florabel. Assuming that aeolian dust material has been uniformly deposited across this landscape, the past and present processes acting at each dune site will be reflected in the degree of preservation of the aeolian material. This includes post-depositional surface processes such as wind and water erosion; if the aeolian deposit has not been adequately stabilised and protected, these processes may have removed and redistributed the deposited dust.

Relating these granulometric data to the parna model of dust transport and deposition, Butler and Hutton (1956) and then Blackburn (1981) asserted that the modal size of the non-clay fraction of deposited dust contributing to parna is widely variable, changing progressively with distance from source. Using low-resolution particle sizing techniques, these authors measured the non-clay fraction mode in parna as being greater than 100 [micro]m diameter within 30km of the assumed source dunes, and 20-60 [micro]m diameter at distances of 150-300 km downwind of the source area. The high-resolution granulometric data from the Hillston sites, plus the similarly detailed granulometric data for deposited dust at Blayney in eastern NSW (Hesse et al. 2003), confirm the likelihood of downwind fining of deposited dust in NSW. At Hillston, some 200 km east and north-east of assumed dust source areas (for parna), the dominant silt modes in the dunes are typically 40-60 [micro]m, whereas at the Blayney site, which is located approximately 350 km due east (and downwind) of the Hillston sites, the dominant silt mode in the soil is distinctly finer at around 30 [micro]m.

Mineralogy

In all phases (s, cs, k) of all dunes (Embagga, Kongong, Florabel), illite and kaolinite dominate the clay-mineral fraction. Smectite is a dominant mineral only in the k phase, a feature that is consistent for all 3 sampling sites (Table 2). Smectite also appears as a minor mineral in the s and cs phases of the Kongong and Florabel dunes, and is absent from these phases in the Embagga dune. Additional minerals present ubiquitously in minor amounts within the clay-size fraction include quartz and primary mica. Interstratified phyllosilicates (smectite-illite) are present in small quantities in most samples.

Studies of modern eastern Australian dust indicate that the dominant clay minerals are illite and kaolinite, and that suspended dusts contain no or only small amounts of smectite (Kiefert and McTainsh 1995). Furthermore, in a review of aeolian dust deposit studies of south-eastern Australia, including various parna studies (e.g. Beattie 1970), Hesse and McTanish (2003) conclude that in the overwhelmingly majority of soils of aeolian origin, illite and kaolinite are the main clay minerals present. The clay mineral suites of the 3 dunes are consistent with an aeolian dust origin for the fines.

In contrast, the dominant clay-sized minerals of the floodplain soil surrounding each of the sand dunes are quartz, smectite, and kaolinite (Cattle et al. 2003). If the fine material sampled from each dune had been wholly derived from the surrounding floodplain, illite would not dominate the dune clay fraction and smectite would comprise a substantial proportion of the clay-mineral suite of all phases. Instead, in most samples, smectite was a subordinate mineral only, with the exception of the k phase of all dunes, the s phase of the Embagga dune, and the sl phase of the Florabel dune. The restricted distribution of smectite within each dune suggests that the smectite is not obtained from the surrounding floodplain, but is either of aeolian origin or has formed pedogenically. Whilst smectite is not generally known to be a major constituent of parna, and is rarely a component of modern-day dust suspensions (Kiefert and McTainsh 1995), its presence in aeolian sediments of eastern Australia has been recorded in a number of studies (Cattle et al. 2002; Chen et al. 2002). It is possible that smectite was deposited on the dunes as a minor constituent of aeolian dust, which subsequently disaggregated, and because of the generally very small size of smectite particles (Pye and Sherwin 1999) these were preferentially illuviated from the upper phases of the dunes.

Alternatively, the smectite in the k phases of the dunes may have formed pedogenically, in situ. Allen and Hajek (1989) suggest that smectite can be formed pedogenically where leaching is restrictive because of low precipitation, or where there is an abundant supply of bases and/or silica. The prevalence of calcium carbonate in the k phase suggests that there would have been an abundant supply of bases available for the neo-formation of smectite.

Micromorphological features

Phase of clay accumulation (cs)

In each of the 3 dunes, the cs phase consists predominantly of detrital mineral grains ranging in diameter from 10 to 1000 [micro]m, plus some fine material. The coarse mineral fraction is overwhelmingly dominated by quartz grains, although there are a number of minor (<5%) mineral components including plagioclase feldspar, epidote, and tourmaline. Occasional concretions of iron oxide and manganese dioxide (20-40 [micro]m diameter) are also present in all samples, and very rare small rutile grains were observed in the Florabel sample. The coarse mineral skeleton grains vary in shape from angular to rounded, are mostly poorly sorted, and of variable sphericity. As shown in Fig. 4, the microstructure of this phase varies among the dunes, but is dominantly bridged grain or compact grain, with some areas of massive structure, depending on the concentration of fine material. In addition to simple, complex, and compound packing voids, some vughs are also present, and floral channels also appear as a minor feature in all samples.

[FIGURE 4 OMITTED]

The elementary fabric of this phase varies between chitonic (coarser units surrounded by a cover of smaller units, indicated by 'C' in Fig. 4), porphyric (large fabric units occur in a dense groundmass of smaller units, indicated by 'P' in Fig. 4), and monic (fabric units of one size, indicated by 'M' in Fig. 4a), a trend related to the concentration and distribution of clay-sized material. The contrasting chitonic and porphyric fabrics are particularly well illustrated in Fig. 4d. A common feature of the cs phase of all dunes is a pattern of layering, where clay, silt, and very fine sand are accumulated in thin bands of porphyric or monic fabric, between which clay accumulation is minimal and a chitonic fabric dominates (Fig. 4a and d). The porphyric fabric is most pronounced in the Florabel dune because of the generally greater clay content, and only in this dune were clay-enriched lamellae visible at the time of sampling.

The clay-sized material in this phase occurs primarily as brown, reddish brown, or yellowish brown coatings (argillans) on larger grains (Fig. 4b and e), or as pore infillings or clay plugs (Fig. 4c). These argillans and infillings represent an important textural pedofeature for all 3 dunes. Sleeman (1975) described similar features in a loessic red-brown earth in northern Victoria as 'wustenquarz', or 'illuviation embedded ferri-argillans'. The argillans here occur primarily as limpid-speckled, continuous, and non-continuous laminated coatings around skeleton grains, packing voids and channels.

Whilst there is some conjecture as to the origin of argillans (a weathering feature or a depositional feature), they are generally thought to result from translocation of clay from upper horizons. As such, argillans are important in providing a record of current and past eluviation-illuviation processes (Bullock et al. 1985). The moderately strongly anisotropic alignment of clay within the argillans, the indiscriminate nature of their distribution, and the micro-laminated structure observed within certain parts (e.g. Fig. 4d) of the cs phase all indicate that the clay has been transported into cs, rather than formed in situ by mineral weathering.

The mobilisation of clay and silt-sized particles and their downward translocation with infiltrating water are among the most common pedogenic processes associated with aeolian dust deposits. Numerous studies have shown that argilluviation of aeolian clay and silt is responsible at least in part for the formation of important pedological features such as argillans (Sleeman 1975), argillic horizons (Beattie 1970), clay lamellae (Holliday and Rawling 2006), and zones of decreased permeability (Chartres 1983). Micromorphological features described here strongly support the presence of an aeolian dust component, with a likely scenario being that the clay material was deposited, disaggregated, or removed from larger grains, and subsequently illuviated lower down the profile. Further, the irregular morphology of the laminated clay layers and the discrete clay lamellae on a macro-scale indicate that direct deposition was not responsible for their formation.

The other micromorphological feature that indicates a prominent aeolian dust component is the abundance of well-sorted 40-70 [micro]m sized quartz grains within the thin, porphyric, or monic bands of all cs phases (e.g. Fig. 4a).

Such grains are easily entrainable by wind during dust storms, and form a relatively impermeable zone at the base of the cs phase of the Embagga and Florabel dunes. The concentration of these coarse silt-sized grains, presumably by illuviation, may have acted as a throttle for the illuviation of later-deposited clay, and may also represent the extent of the wetting front.

Phase of carbonate accumulation (k)

The soil of the k phase consists predominantly of detrital mineral grains ranging in diameter from 10 [micro]m to 1000 [micro]m, and some scarce fine material. As in the cs phase, the coarse mineral fraction of the k phase in all dunes is dominated by quartz grains, which are typically 80-200 gm in diameter. Minor (<5%) mineral components in the coarse grain fraction include zircon, orthoclase feldspar, and subrounded grains of hornblende. The coarse mineral grains show variable sphericity, are dominantly angular to rounded, and are poorly sorted. The microstructure is relatively consistent within this phase of each dune, being dominantly pellicular grain and bridged grain. Irregular nucleic concretions (glaebules) composed of coarse mineral grains cemented with crystalline calcite appear consistently as a pedofeature in this phase of each dune (Fig. 5a). Porosity is dominated by compound packing voids, as well as relatively rare floral channels. Close-packed bridge structure predominates within the calcite glaebules.

[FIGURE 5 OMITTED]

The fine mineral material in all k samples consisted chiefly of calcite, with some brown to reddish clay and iron-oxide also present. Both the clay and the iron-oxides are present largely as thin, incomplete coatings on, and bridges between, detrital grains.

Extreme birefringence and distinctive interference colours make secondary micritic calcite readily observable in thin sections (Fig. 5b and c). Here calcite is present as dense crystals in abundant glaebules where it forms micritic hypocoatings that cement quartz grains of varying size (10-100 [micro]m) and shape (spherical, subspherical, and tabular). The shapes of the concretions are irregular, although dominantly subspherical, and size shows substantial variation within the range 10-3000 [micro]m. Whilst present at all sites, calcite glaebules are most prevalent in the k phase of Florabel, and in the lower part of the Kongong k phase. In the top part of the Kongong k phase, calcite was often present as incomplete coatings, or fragments of coatings, rather than as discrete glaebules.

The elementary fabric of this phase is monic or chitonic outside of the calcite glaebules, where fine material is relatively scarce, and porphyric within the glaebules, where calcite crystals form a groundmass of varying density. Figure 5c illustrates the contrasting fabrics within the k phase of the Florabel dune. The fabric of smaller glaebules is frequently close-porphyric, and pore space generally increases with the size of the glaebule.

Calcitic pedofeatures are formed when deposited carbonates are dissolved, distributed throughout the soil profile with infiltrating water, and re-precipitated as coatings on existing grains, voids, or plant structures, and as void or root-channel infillings (Bullock et al. 1985). Although the presence of calcium carbonate is not diagnostic of parna profiles, numerous studies of parna and loessic soils of south-eastern Australia have identified subsoil zones of diffuse secondary carbonate accumulation (Beattie 1970; Sleeman 1975; Chartres 1983). Butler's (1956) original model of parna included a large range of calcium carbonate content (5-25%), varying inversely with distance from the source region and annual rainfall. Micromorphological features, including the fine crystalline nature of the calcite, which indiscriminately coats detrital grains, suggest that the accumulation of calcite in the k phase of each of the dunes is the result of a re-distribution of deposited calcium carbonate. This material is likely to be derived from an allochthonous source, as there is no abundant local topsoil source of calcium carbonate. However, the origin of the allochthonous calcium carbonate may not necessarily be only terrestrial dust, as Quade et al. (1995) demonstrated that secondary soil carbonates may also form through the combination of calcium ions dissolved in rainfall and plant-derived carbon in the soil. The contribution of deposited dust to the calcite content of these dunes therefore remains unclear.

Grain morphology

A clear morphological feature of many grain samples from the cs phase of each dune was the presence of a thin coating of fine material covering the silt and sand-sized quartz particles (e.g. centre grain, Fig. 6a). These argillans further indicate that illuviation has been an important post-depositional weathering process in these dunes; some of the quartz grains examined by SEM were bridged by thin, smooth, laminated argillans (Fig. 6b), suggesting the influence of percolating water, rather than in situ weathering of coarse mineral grains. As indicated in Fig. 6c and d, a substantial proportion of these quartz grains were 30-70 [micro]m in diameter, which is typical for wind-transported particles. The shape of these grains showed some variation both within and between all samples. Some, such as those pictured in Fig. 6e and f, were conspicuously angular, whereas others, such as those pictured in Fig. 6a and b, were subrounded.

[FIGURE 6 OMITTED]

The shape of aeolian grains is an issue of some conjecture; several authors suggest that active aeolian sand grains can vary from subrounded to subangular (Stapor et al. 1983; Tchakerian 1999). Pye and Sherwin (1999) suggest that the shape of aeolian dust grains is a reflection of mineral composition, mode of formation, and the degree of pre- and post-depositional modification induced by weathering. Dust grains derived from secondary sources, such as the aeolian dust described in south-eastern Australia, may tend to be more rounded due to the coatings of clay-sized minerals; the subrounded 30-70 [micro]m quartz grains evident in Fig. 6a and b are good examples of this, but other angular grains of the same size indicate that shape is not necessarily a useful diagnostic criterion for aeolian dust provenance.

Dust-mediated pedogenesis in the 3 sand dunes and the question of parna

The nature and the distribution of fine-grained mineral material 'trapped' within the 3 source-bordering sand dunes at Hillston suggest that it has not been obtained locally. As the sand dunes rise well above the level of the floodplain they are unlikely to have received fluvial deposits, indicating that any fine material present will have been a result of either aeolian deposition, or weathering of coarse mineral grains in situ. The consistency of mineral suite and the lack of weatherable minerals in the sand-sized fraction of all 3 dunes suggest that aeolian accessions are the more likely source of this material. The clay mineral suite of the upper 2 dune phases is dominated by illite and kaolinite, whereas smectite is consistently only a dominant mineral in the k phase of each dune. As the surrounding floodplain soils are rich in smectite and comparatively poor in illite (Cattle et al. 2003), an allochthonous origin appears certain.

The relative consistency of morphological features of the 3 dunes indicates that they have been subject to similar deposits of dust and subsequent pedogenic processes. Although there is some variation in structure, fabric, and texture grade, the distinctive reddish, clay-enriched phase (cs) present in each dune is interpreted as illuviated aeolian dust. The l phase soil of the Florabel dune represents a particularly concentrated variant of this illuviated material. Micromorphological examination of the cs and l phases indicates that the clay present appears as strongly anisotropic argillans, which are known to be a common illuvial pedofeature (Bullock et al. 1985). Furthermore, the laminated arrangement of fine (30-70 [micro]m) quartz grains within the cs phase suggests that this material may have been illuviated with or before the aeolian clay, possibly forming a relatively impermeable layer that impeded further particle translocation. Red clayey bands in dune sequences found in New South Wales and other parts of the world have similarly been interpreted as illuviated aeolian dust (e.g. Jorgensen 1992; Page et al. 2001; Chen et al. 2002; Holliday and Rawling 2006).

The calcium carbonate glaebules abundant in the k phase of each dune are another feature that may be linked to the aeolian dust component. The fine crystalline nature of the calcite, which has infilled various pores and ubiquitously coats the matrix mineral grains of each lower dune phase, suggests that this material may have been deposited as a constituent of dust, solubilised and translocated from surface horizons, and then re-precipitated at depth. Secondary calcareous deposits have been commonly associated with aeolian deposits across south-eastern Australia (Page et al. 2001); as an example, Beattie (1970) described an accumulation of secondary calcite deposits in parna subsoils of southern New South Wales. However, alternative mechanisms of soil carbonate development have been demonstrated in semi-arid Australia (e.g. Quade et al. 1995), and so the k phase glaebules cannot be regarded as the strongest indicators of aeolian dust-mediated pedogenesis in these dunes.

Whilst it is clear that the fine material in these dunes is of aeolian origin, it is not clear whether this material was deposited as aggregated clay pellets, as proposed in Butler's (1956) parna model. The mineral constituents that have been added to the dunes, silt-sized quartz, clay, and calcium carbonate, are all assumed components of parna, but there is no micromorphological evidence of pelletal aggregations of these materials in any of the dune phases. The only consistent 'aggregations' to appear are those of quartz grains coated in thin clay coats (cs phase) or calcite coats (k phase), both of which are secondary features. This is confirmed by the high-resolution granulometric data; for most of the samples analysed, only small amounts of clay are liberated by complete dispersion of samples when a comparison of the minimally and fully dispersed particle size distributions is made. Although Butler himself (Butler 1974) was unsure of the resilience of parna aggregates once deposited, Blackburn (1981) found that parna can be difficult to fully disperse, intimating that this material should be long-lived. There is no evidence of long-lived clay aggregates in the Hillston dunes.

It is therefore more appropriate to regard the fine material in the Hillston dunes as simply the re-distributed breakdown products of clayey desert dust. Although it is impossible to quantify the clay content of the deposited dust due to subsequent size-sorting and re-distribution of dust components throughout the dunes, the wind-transportable, non-sand fractions of various dune layers contain appreciable clay. In the cs phases of the dunes, the clay content of the non-sand fraction (<75 [micro]m) varies from 34 to 47%, whereas the non-sand fraction of the l phase of the Florabel dune contains 87% clay. Even in the overwhelmingly sandy s phases of the dunes, the non-sand fraction contains 45-78% clay. Only in the sl phase of the Florabel dune and the k phase of the Kongong dune do the non-sand fraction clay contents fall below 30%. What remains unclear is whether the clay was transported as coatings on silt-sized quartz grains, of which there are many in the cs phases of the dunes, or as silt-sized aggregations of clay suspended in wind with similarly sized 'companion' quartz grains.

Conclusions

Many Australian soils contain an aeolian dust component, but the characteristics of pure aeolian materials and the true extent of their distribution remain unclear. To confirm the aeolian provenance of fine-grained material trapped in 3 source-bordering sand dunes on the lower Lachlan floodplain, an upper-dune sandy phase, a mid-dune clay-enriched phase, and a lower-dune carbonate-rich phase were characterised.

Granulometric analyses revealed an accumulation of both clay and a conspicuous coarse silt-sized particle population (20-60 [micro]m diameter) in the clay-enriched phase of each dune. The clay mineral suite of the upper-dune phases is dominated by illite and kaolinite, which is consistent with the known composition of both suspended aeolian dust and aeolian dust deposits of south-eastern Australia. Smectite, which dominates the surrounding floodplain soil, is a dominant mineral in only the lower phase of all dunes, and may have accumulated there through preferential illuviation from the upper phases, or neo-formation in situ. Micromorphological characteristics such as the extensive and indiscriminate distribution of argillans, clay plugs, and wustenquarz throughout the mid-dune phase of all dunes suggest that argilluviation has been a dominant pedological process, and discredits the possibility that this clay has weathered in situ from matrix mineral grains. The abundant silt-sized quartz grains in the clay-enriched dune phases frequently occurred in a wavy, laminated structure, suggesting that these grains may have been translocated downwards with infiltrating water, perhaps concurrently with the clay particles. A similar process of translocation appears to be responsible for the occurrence of fine-grained crystalline calcite that bridges matrix mineral grains in the lower phase.

The consistency of morphological and mineralogical features for all 3 dunes supports the hypothesis of an analogous dust accession, and suggests that similar soil-forming processes have taken place, namely the translocation and transformation of this deposited aeolian dust. The dominant model for aeolian dust in south-eastern Australia is 'parna', a clayey, calcareous material thought to have been transported as silt-sized aggregates. Whilst no discrete aggregates were identified, all the assumed components of parna (silt-sized quartz grains, clay, and calcium carbonate) were identified as separate entities, spatially separated within each dune. It would appear that these source-bordering sand dunes have acted as 'dust traps' and contain the breakdown products of clayey desert dust. Given the nature of aeolian processes, soils of the entire region are likely to have received inputs of this dust. The subsequent in situ alteration of this dust is likely to have played an important role in the pedogenesis of the agriculturally important soils of the surrounding lower Lachlan valley.

Manuscript received 28 April 2006, accepted 20 October 2006

References

Allen BL, Hajek BF (1989) Mineral occurrence in soil environments. In 'Minerals in soil environments'. (Eds JB Dixon, SB Weed) pp. 99 278. (Soil Science Society of America: Madison, WI)

Beattie JA (1970) Peculiar features of soil development in parna deposits in the Eastern Riverina, NSW. Australian Journal of Soil Research 8, 145-156. doi: 10.1071/SR9700145

Blackburn G (1981) Particle-size analyses of Widgelli parna in southeast Australia. Australian Journal of Soil Research 19, 355-360. doi: 10.1071/SR9810355

Brindley GW, Brown G (1980) X-ray diffraction procedures for clay mineral identification. In 'Crystal structures of clay minerals and their X-ray identification'. (Eds GW Brindley, G Brown) pp. 305 361. (Mineralogical Society: London)

Bullock P, Fedoroff N, Jongerius A, Stoops G, Tursina T (1985) 'Handbook for soil thin section description.' (Waine Research: Albrighton, UK)

Bureau of Meteorology (1988) 'Climatic averages Australia.' (Australian Government Publishing Service: Canberra) Butler BE (1956) Parna--an aeolian clay. The Australian Journal of Science 18, 145-151.

Butler BE (1974) A contribution towards the better specification of parna and some other aeolian clays in Australia. Zeitschrift fur Geomorphologie N.F. Suppl. 20, 106-116.

Butler BE, Hutton JT (1956) Parna in the Riverine Plain of south-eastern Australia and the soils thereon. Australian Journal of Agricultural Research 7, 536-553. doi: 10.1071/AR9560536

Cattle SR, McTainsh GH, Wagner S (2002) AEolian dust contributions to soil of the Namoi Valley, northern NSW, Australia. CATENA 47, 245-264. doi: 10.1016/S0341-8162(01)00181-3

Cattle SR, Meakin SN, Ruszkowski P, Cameron RG (2003) Using radiometric data to identify aeolian dust additions to topsoil of the Hillston district, western NSW. Australian Journal of Soil Research 41, 1439-1456. doi: 10.1071/SR03026

Chartres CJ (1983) The pedogenesis of desert loam soils in the Barrier Range, western New South Wales. II. Weathering and soil formation. Australian Journal of Soil Research 21, 1-13. doi: 10.1071/SR9830001

Chen XY (2001) The red clay mantle in the Wagga Wagga region, New South Wales: evaluation of an aeolian dust deposit (Yarabee Parna) using methods of soil landscape mapping. Australian Journal of Soil Research 39, 61-80. doi: 10.1071/SR99105

Chen XY, Spooner NA, Olley JM, Questiaux DG (2002) Addition of aeolian dusts to soils in southeastern Australia: red silty clay trapped in dunes bordering Murrumbidgee River in the Wagga Wagga region. CATENA 47, 1-27. doi: 10.1016/S0341-8162(01)00176-X

Dare-Edwards AJ (1984) AEolian clay deposits of south-eastern Australia: parna or loessic clay? Transactions of the Institute of British Geographers 9, 337-344. doi: 10.2307/622237

Gee GW, Bander JW (1986) Particle-size analysis. In 'Methods of soil analysis: Part 1. Physical and mineralogical methods'. (Ed. A Klute) pp. 383-411. (American Society of Agronomy Inc.: Madison, WI)

Hesse PP, Humphreys GS, Smith BL, Campbell J, Peterson EK (2003) Age of loess deposits in the Central Tablelands of New South Wales. Australian Journal of Soil Research 41, 1115-1131. doi: 10.1071/SR02152

Hesse PP, McTanish GH (2003) Australian dust deposits: modem processes and the Quaternary record. Quaternary Science Reviews 22, 2007-2035. doi: 10.1016/S0277-3791(03)00164-1

Holliday VT, Rawling JE (2006) Soil-geomorphic relations of lamellae in eolian sand on the High Plains of Texas and New Mexico. Geoderma 131, 154-180. doi: 10.1016/j.geoderma.2005.03.019

Isbell RF (1996) 'The Australian soil classification.' (CSIRO Publishing: Melbourne, Vic.)

Jorgensen DW (1992) Use of soils to differentiate dune age and to document spatial variation in eolian activity, northeast Colorado, USA. Journal of Arid Environments 23, 19-34.

Kiefert L, McTainsh GH (1995) Clay minerals in east Australian dust. In 'Clays, controlling the environment: Proceedings of the 10th International Clay Conference'. Adelaide, Australia, 1993. (Eds GJ Churchman, RW Fitzpatrick, RA Eggleton) pp. 339-345. (CSIRO Publishing: Melbourne, Vic.)

McKenzie N, Jacquier D, Isbell R, Brown K (2004) 'Australian soils and landscapes: an illustrated compendium.' (CSIRO Publishing: Melbourne, Vic.)

McTainsh GH, Lynch AW, Hales R (1997) Particle-size analysis of aeolian dusts, soils and sediments in very small quantities using a Coulter Multisizer. Earth Surface Processes and Landforms 22, 1207-1216. doi: 10.1002/(SICI)1096-9837(199724)22:13<1207:: AID-ESP820>3.0.CO;2-K

Page KJ, Dare-Edwards AJ, Owens JW, Frazier PS, Kellett J, Price DM (2001) TL chronology and stratigraphy of riverine source bordering sand dunes near Wagga Wagga, New South Wales, Australia.

Quaternary International 83--85, 187-193. doi: 10.1016/S10406182(01)00039-8 Pye K (1987) 'AEolian dust and dust deposits.' (Academic Press: London)

Pye K, Sherwin D (1999) Loess. In 'AEolian Environments, sediments and landforms'. (Eds AS Goudie, I Livingstone, S Stokes) pp. 213-238. (John Wiley and Sons Ltd: Chichester, UK)

Quade J, Chivas AR, McCulloch MT (1995) Strontium and carbon isotope tracers and the origins of soil carbonate in South Australia and Victoria. Palaeogeography, Palaeoclimatology, Palaeoecology 113, 103-117. doi: 10.1016/0031-0182(95)00065-T

Sleeman JR (1975) Micromorphology and mineralogy of a layered red-brown earth profile. Australian Journal of Soil Research 13, 101-119. doi: 10.1071/SR9750101

Stapor FW, May JP, Barwis J (1983) Eolian shape shorting and aerodynamic traction equivalence in the coastal dunes of Hout Bay, Republic of South Africa. In 'Developments in sedimentology: eolian sediments and processes'. (Eds ME Brookfield, TS Ahlbrandht) pp. 149-164. (Elsevier Science Publishers B.V: Amsterdam)

Tchakerian VP (1999) Dune paleoenvironments. In 'AEolian environments, sediments and landforms'. (Eds AS Goudie, I Livingstone, S Stokes) pp. 261-292. (John Wiley and Sons Ltd: Chichester, UK)

Walker PH, Chartres CJ, Hutka J (1988) The effect of aeolian accessions on soil development on granitic rocks in south-eastern Australia. I. Soil morphology and particle-size distributions. Australian Journal of Soil Research 26, 1-16.

Wasson RJ, Fitchett K, Mackey B, Hyde R (1988) Large-scale patterns of dune type, spacing and orientation in the Australian continental dunefield. Australian Geographer 19, 89-104.

Adrienne L. Ryan (A) and Stephen R. Cattle (A, B)

(A) Sciences Discipline, Faculty of Agriculture, Food and Natural Resources, The University of Sydney, NSW 2006, Australia.

(B) Corresponding author. Email: s.cattle@usyd.edu.au
Table 1. Physical attributes of the various phases of the 3 dunes

Structural condition is summarised by the degree of pedality and
the ped shape: wp, weakly platy; am, apedal massive; asg, apedal
single grained

Thickness Phase name Field texture Structure
(m) grade

 Embagga

2 Coarse sand (s) Sand wp
1 Clayey sand (cs) Sandy loam am
1.5 Calcareous sand (k) Sand asg

 Kongong

1.5 Coarse sand (s) Sand asg
1 Clayey sand (cs) Fine sandy loam wp
1.5 Calcareous sand (k) Sand asg

 Florabel

2 Coarse sand (s) Sand asg
1 Clayey sand (cs) Sandy loam am
0.01-0.04 Lamellae (l) Sandy clay loam am
0.15 Sublamellae (sl) Sandy loam am
2.5 Calcareous sand (k) sand asg

Thickness Phase name Dry colour Other features
(m)

 Embagga

2 Coarse sand (s) 10YR 6/6 --
1 Clayey sand (cs) 5YR 5/6 Medium to coarse sand
1.5 Calcareous sand (k) 10YR 6/4 Frequent cylindrical
 glaebules (5-20 mm
 diameter, 10-70 mm
 length) comprising
 calcite and coarse
 sand grains

 Kangong

1.5 Coarse sand (s) 10YR 6/6 --
1 Clayey sand (cs) 7.5YR 5/6 Fine to medium sand
1.5 Calcareous sand (k) 10YR 6/4 Some cylindrical
 glaebules (5-15 mm
 diameter, 10-20 mm
 length) comprising
 calcite and coarse
 sand grains

 Florabel

2 Coarse sand (s) 10YR 6/6 --
1 Clayey sand (cs) 5YR 5/8 Medium to coarse sand
0.01-0.04 Lamellae (l) 5YR 5/8 Medium to coarse sand
0.15 Sublamellae (sl) 10YR 6/6 Fine to medium sand
2.5 Calcareous sand (k) 10YR 7/4 Frequent cylindrical
 glaebules (5-10 mm
 diameter, 10-30 mm
 length) comprising
 calcite and coarse
 sand grains; some
 calcified plant roots

Table 2. The dominant and subdominant minerals present in the
clay-size fraction of soil from the 3 dune sampling sites

Dune phase Sampling site Dominant minerals

s Embagga Illite, kaolinite, smectite
 Kongong Illite, kaolinite
 Florabel Illite, kaolinite

cs Embagga Illite, kaolinite
 Kongong Illite, kaolinite
 Florabel Illite, kaolinite

l Florabel Illite, kaolinite

sl Florabel Illite, kaolinite, smectite

k Embagga Illite, kaolinite, smectite
 Kongong Illite, kaolinite, smectite
 Florabel Illite, kaolinite, smectite

Dune phase Sampling site Subdominant minerals

s Embagga Quartz, interstratified minerals, mica
 Kongong Smectite, quartz, interstratified
 minerals, mica
 Florabel Smectite, quartz, interstratified
 minerals, mica

cs Embagga Quartz, mica
 Kongong Smectite, quartz, interstratified
 minerals, mica
 Florabel Smectite, quartz, interstratified
 minerals, mica

l Florabel Quartz, interstratified minerals, mica

sl Florabel Quartz, interstratified minerals, mica

k Embagga Quartz, interstratified minerals, mica
 Kongong Quartz, interstratified minerals, mica
 Florabel Quartz, interstratified minerals, mica,
 vermiculite
COPYRIGHT 2006 CSIRO Publishing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2006 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Ryan, Adrienne L.; Cattle, Stephen R.
Publication:Australian Journal of Soil Research
Geographic Code:8AUST
Date:Dec 1, 2006
Words:7545
Previous Article:Prediction of spatial ET-fluxes using remote sensing and field data of selected areas in the eastern part of Abu Dhabi, United Arab Emirates.
Next Article:Unravelling the effects of soil properties on water infiltration: segmented quantile regression on a large data set from arid south-west Africa.
Topics:


Related Articles
The sandy face of Mars: astronomers ponder the awesome dimensions of Martian dunelands.
Sandy points.
Runoff and micromorphological properties of a grazed haplargid, near Cobar, NSW, Australia.
Martian sand dunes: blowing in the wind.
Modelling current parna distribution in a local area.
Symposium report: Aeolian dust--implications for Australian mineral exploration and environmental management.
The red clay mantle in the Wagga Wagga region, New South Wales: evaluation of an aeolian dust deposit (Yarabee Parna) using methods of soil landscape...
Shifting sands: dunes hold heaps of material--and climate clues, too.
Using radiometric data to identify aeolian dust additions to topsoil of the Hillston district, western NSW.
Rhythms of the desert: physicists are starting to unravel the mystery of dunesong produced by the Kelso Dunes in Mojave National Preserve.

Terms of use | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters