Printer Friendly

Iron nodules in ferric soils of the Fraser Coast, Australia: relicts of laterisation or features of contemporary weathering and pedogenesis?

Introduction

Iron (Fe) nodules and pisoliths are a common feature of many soils and weathering profiles in the tropics, subtropics, regions of Mediterranean climate and even temperate regions (Amouric et al. 1986; Milnes et al. 1987; Tardy 1992; Bourman 1993a; Singh and Gilkes 1996; Breuning-Madsen et al. 2007; Schulz et al. 2010). Numerous studies have been published, mostly focusing on nodule microstructure, but also their chemistry and mineralogy (Childs 1975; Anand and Gilkes 1987; Milnes et al. 1987; Tardy 1997; Latrille et al. 2001; Gasparatos et al. 2004; Tripathi and Rajamani 2007). In most cases, Fe nodules are composed of quartz grains cemented by various Fe and aluminium (AI) oxyhydroxides.

Soils or soil horizons which contain a high proportion of ferruginous nodules are widespread in Australia, where they are classified as 'ferric' soils (lsbell 2002). Several modes of formation have been proposed to explain the occurrence and distribution of Fe nodules in soils and weathering profiles of coastal eastern Australia, such as breakdown of laterite duricrnsts; weathering of pre-existing, Fe-rich formations; or redox-controlled formation. However, the mode of formation of these nodules continues to be a source of debate (Bourman 1993a, 1993b; Schulz et al. 2010) and it is likely that Fe nodules found in different parts of Australia have differing genetic pathways (e.g. Singh and Gilkes 1996; Anand 2001; Anand and Paine 2002; Taylor and Eggleton 2008).

Laterisation refers to the in-situ formation of an Fe-rich duricrust capping a deep weathering profile (Tardy 1992). The laterite duricrust is a product of prolonged chemical weathering and forms by relative accumulation of Fe and A1 as more soluble components are lost from the weathering profile. Laterisation is thought to require subdued topography and a humid tropical or subtropical climate (Widdowson 2008). It is considered to be an important process in parts of Africa, South America, the Indian Subcontinent, and some parts of Australia (Tardy 1992; Anand and Paine 2002; Taylor and Eggleton 2008; Widdowson 2008). Tardy (1992) considered the breakdown of laterite duricrust as the result of continuing surficial weathering to be an important pathway for Fe nodule formation, and this mode of formation has also been proposed for nodules from the Yilgarn Craton of Western Australia (Anand and Paine 2002). Similarly, the widely occurring Fe nodules and duricrusts in coastal eastern Australia have been described as eroded remnants of laterite duricrusts (e.g. Faniran 1970; Cranfield 1994) and are thought to have formed during a period of tropical weathering in the Miocene.

In periodically saturated soils, Fe nodules can form in the zone of watertable fluctuation (Blume 1968; Schwertmann and Fanning 1976). Iron from sparely soluble Fe oxides is released under the anoxic, reducing conditions commonly encountered in saturated soils. The Fe nodules then form by the re-precipitation of Fe oxides as a result of seasonal changes to the soil redox potential, infilling the soil pores and cementing the soil matrix (Singh and Gilkes 1996; White and Dixon 1996; Zhang and Karathanasis 1997). Nodule size and abundance generally increase with greater duration of seasonal saturation (Schwertmann and Fanning 1976). This mechanism has been used to explain the occurrence of some Fe nodules in soils from coastal south-western Australia (Singh and Gilkes 1996) and northern Queensland (Coventry et al. 1983) and the distribution of Fe nodules in the soils of coastal south-east Queensland (Coaldrake 1961).

Iron nodules have also been shown to form as the result of the gradual breakdown of material such as Fe-rich sandstones (Hunt et al. 1977) or ferricrete or Fe-mottles (Bourman 1993a). Where these materials occur at elevated topographic positions, they can be dissected, eroded, and transported downslope by colluvial processes. The eroded fragments continue to be altered by weathering and are rounded by abrasion during transport. In some instances, Fe nodules may grow by accretion of additional Fe cement. Where multiple cycles of erosion, deposition, and accretion occur, this may result in nodules with a characteristic concentric morphology (Milnes et al. 1987; Bourman 1993a). This mode of formation is distinguished from laterisation as it does not require long periods of weathering under tropical or subtropical conditions and can occur in a broader range of lithologies. Bourman and co-workers (Milnes et al. 1985, 1987; Bourman 1993a, 1993b) have demonstrated that this process is responsible for the formation of Fe nodules in South Australia and South Africa, while Hunt et al. (1977) used it to explain the distribution and morphology of Fe nodules in the Sydney region of south-eastern Australia.

The aim of this study is to determine the genetic mechanisms of the ferric horizons and Fe nodules commonly found in the soils of the Fraser Coast of south-east Queensland (Fig. 1). To determine the mode of formation of these Fe-rich features, we establish the distribution of Fe nodules in Poona Creek catchment and the field relationships, micromorphology, and mineralogy of a ferric soil profile exposed in an 'ironstone gravel' quarry. Our study also presents detailed descriptions and images.

Regional setting

Location and climate

Poona Creek catchment is on the Fraser Coast of Queensland, Australia (Fig. 1). The region has a subtropical climate, with an average annual rainfall of 1270 ram, occurring mostly from October to March. Mean monthly maximum temperatures range from 30.2 to 21.5[degrees]C, in December and July, respectively (Australian Bureau of Meteorology, www.bom.gov.au/climate/averages/tables/cw_040126.shtml).

Geomorphology and geology

Although elevation reaches 120 m above sea level in the west of the catchment, it is mostly <30 m and decreases gently towards the coastline in the east. Most of the area is of low gradient, although there are steeper regions in the elevated, western part. Poona Creek, which drains the catchment, is tidal in its lower estuarine section and discharges into the Great Sandy Strait.

Bedrock in this region consists of mudstone, shale, siltstone, and sandstone of the Duckinwilla Group, and contains several Fe-rich units (Cranfield 1994). These Late Triassic--Early Jurassic age sedimentary rocks are overlain by the Late Jurassic-Early Cretaceous volcanic Grahams Creek Formation to the east of the study area and remnants of the fluvial Elliott Formation to the north. It has been suggested that these formations were deeply weathered by a 'laterisation episode' in the Miocene (Cranfield 1994).

Soils

The soils in the study area are varied. A sequence of Red Kandosols and Kurosols (Australian Soil Classification; lsbell 2002) (World Reference Base equivalents (IUSS Working Group WRB 2006)): Rhodic Cambisols and Acrisols on hilltops, Red and Yellow Kandosols or Kurosols (often ferric) at midslope to footslope positions (WRB equivalents: Rhodic/ Xanthic Cambisols and Acrisols), and (ferric) Redoxic Hydrosols or Semi-Aquic Podosols (WRB equivalents: Gleyic Luvisols and Endogleyic Podzols) in poorly drained valleys and on creek margins has been observed (Coaldrake 1961). Seasonal waterlogging is a common feature in lower gradient zones across much of the catchment. Soils commonly feature a ferric horizon, which can reach thicknesses > l m.

Methods

Sample collection

Soil samples (120, from top 30 cm of soil) were collected across the Poona catchment to map the distribution of Fe nodules. Sample locations were determined by dividing the catchment using a grid pattern and collected at random within individual grid cells. The percentage by weight of nodules in each sample was recorded after passing the air-dried samples through a 2 mm sieve.

In addition, a 2.5 m deep ferric soil profile exposed at a small ironstone gravel quarry (Grid Reference (i.e. GDA coordinates): 0485856 E, 7145385 N) was selected for more detailed study. The quarry is one of many small quarries established to excavate Fe nodules for use as road-base material on plantation roads. Although anthropogenic modification of the profile cannot be completely ruled out, field evidence and typical quarrying practices suggest that the profile is undisturbed and representative of ferric soils in the region. At least one bulk sample was collected from each soil horizon. Oriented samples for preparation of polished blocks and thin sections were collected where possible. Only samples from the ferric soil profile were used for further analyses.

[FIGURE 1 OMITTED]

Sample processing

Bulk samples collected from the profile were sieved into five size fractions (<2, 2-4, 4 8, 8 16, and >16mm) to separate rock clasts and Fe nodules of different sizes from the fine earth fraction. The nodules were further separated into magnetic and non-magnetic fractions using a neodymium hand magnet. The weight of each fraction was recorded.

Subsamples of each size fraction were milled into a fine powder for quantification of bulk mineralogy. Prior to separation of magnetic and non-magnetic nodules, 50 nodules from each size fraction of all samples (1400 total) were selected for macromorphological description (roundness, sphericity, appearance of the outer cortex, hardness, and internal morphology). Soil thin sections for micromorphological description were prepared from oriented samples following standard procedures.

Mineral phase identification and quantification

Bulk mineralogy of the samples was determined using X-ray diffractometry (XRD). Samples were milled to a fine powder in an agate swing mill, and reduced to a 5-1am powder in McCrone Micronizing Mills (McCrone Microscopes & Accessories, Westmont, IL), using agate beads and ethanol. The XRD analysis was performed on the micronised powders (+ 10% corundum as an internal standard) using an X'Pert PRO Multi Purpose Diffractometer (PANalytical B.V., Almelo, The Netherlands) with a copper anode, followed by identification and quantification of the crystalline material using Jade 9 (Materials Data, Inc., Livermore, CA) and SiroQuant (CSIRO), respectively. The internal standard permits quantification of poorly diffracting materials (PDM) and determination of absolute mineral phase concentrations using Rietveld refinement in SiroQuant. All X-ray diffractograms were clustered using the whole-pattern hierarchical clustering capabilities of PANalytical HighScore Plus software. This permits rapid identification of similar samples and thus of mineralogy trends with nodule size and sample depth.

Iron oxides in thin section were distinguished using colour (Scheinost and Schwertmann 1999) and Raman microspectroscopy (Hanesch 2009). A Raman 3000 microscope system (Renishaw plc, Wotton-Under-Edge, UK) equipped with a 1200 lines/mm grating and a charge-coupled device (CCD) detector was used. This system incorporates a Leica optical microscope with 20 x and 50 x objective lenses. Spectra were excited by the 785 nm line of a Renishaw diode laser operating at 0.2 mW laser power.

Aluminium substitution in Fe oxides

In addition to quantifying individual mineral phases, Fe-oxide unit cell sizes obtained by Rietveld refinement in SiroQuant were used to calculate the degree of Al-for-Fe substitution in the Fe-oxide phases. Linear relationships between Al substitution and unit cell size have been found for both natural maghemites (Schwertrnann and Fechter 1984) and natural goethites (Schwemnann and Carlson 1994) from tropical soils. As the relationship between the extent of AI substitution and the unit cell parameters of soil hematites has not been established to date, the degree of substitution was calculated using the relationship between unit cell size and Al substitution derived from Al-substituted hematites synthesised under conditions analogous to the soil environment (Schwertmann et al. 2000). Aluminium substitution in Fe oxides from the fine earth fraction and goethite in the magnetic nodules was not determined, as mineral concentrations were too low (<6%).

Micromorphology

Schulz et al. (2010) noted that nodule morphology is generally overlooked as an attribute although it is important for understanding the nodule-formation process. Bourman (1993a, 1993b) also emphasises that different modes of formation result in distinct external and internal morphologies. We have tried to address these shortcomings, and nodule micromorphology has been examined in considerable detail in an effort to clarify nodule genesis.

Results

Iron nodules are a common component of the surficial soils of the study area. They constitute up to 47 wt% of surficial soils, but are more commonly in the range 0-3 wt% (Fig. 2). Soils with greater amounts of Fe nodules occur in the more steeply sloped, south-western parts of the study area; very few were observed in the soils of the coastal plain.

Ferric soil profile

The ferric soil profile selected for detailed investigation is in a mid-slope position. A sketch of the profile is shown in Fig. 3 and a detailed description is given in Table 1. The soil is a Ferric, Brown Kandosol (Australian Soil Classification; Isbell 2002). Main features of the profile include the platy, Fe-nodule-rich upper horizon (A1), with a sharp boundary to the massive, largely nodule-free horizon below (A2; see Fig. 3a). The quantity and size of the Fe nodules then increase with depth, reaching a maximum in the lower part of the clast-supported, ferric, B2 horizon (>50%; Figs 3 and 4). The B2 horizon is underlain by a horizon containing numerous, large ferruginous sandstone clasts (CB) and a C horizon which contains smaller, less Fe-rich sandstone fragments in a medium clay matrix. Magnetic nodules are confined to the parts of the profile above [approximately equal to] 120cm and are most abundant in B1 and upper B2 horizons (Fig. 4). There is evidence of bioturbation in the A1, A2, and B1 horizons, mostly in the form of active and infilled ant tunnels (krotovinas). Surface ant mounding is common in the area.

Nodule morphology

Unbroken nodules are light to dark brown and magnetic nodules are generally darker. The outer surfaces of both magnetic and non-magnetic nodules in the top horizons (excluding A1) have a smooth, burnished appearance. Non-magnetic nodules of the A1 horizon, and those below the upper B2 horizon, have a rougher external morphology in which quartz grains protrude from the nodule surface. Nodules from A1, A2, BI, and upper B2 horizons are well rounded or rounded. However, angularity increases sharply at ~1.5 m depth. All nodules are indurated and cannot be hand-crushed.

Several morphotypes were distinguished in both magnetic and non-magnetic nodules. Representative examples of these are shown in Fig. 5 and their distinguishing properties are listed in Table 2. Non-magnetic nodules are surprisingly porous considering the force required to break them open. However, a cortex that is harder and less porous than the core is a characteristic common to all non-magnetic nodules >4 mm in diameter. Overall, uniform/gradational nodule morphotypes are most common in the smaller nodules (2 8 ram), whereas concentric and nucleic nodules are more commonly >8 mm. Magnetic nodules (mainly <8 ram) have less complex internal morphology and most are of the low porosity, massive morphotype. A subset of magnetic nodules has intermediate properties between massive and porous magnetic morphotypes, i.e. a glassy, shiny core surrounded by a more porous, reddish brown matrix, or vice-versa.

Nodule morphotype changes markedly with depth, and also as a function of nodule size (Fig. 6). The greatest proportion of concentric and nucleic morphotypes occurs in the B2 horizon. Nodules in this horizon also show the greatest number of distinct morphological zones (data not shown). Non-magnetic nodules <8 mm are most likely to be of the uniform/gradational type, regardless of depth. Massive magnetic nodules are most common in the 4-8 mm size fraction of the A l, B1, and B2 horizons, whereas the porous magnetic type occurs more frequently in other horizons and size fractions. Iron-rich sandstone clasts are common in the CB/C horizons, and several were observed in the >16mm fraction of the A1 horizon (Fig. 6).

Mineralogy of Fe nodules, sandstone clasts, and fine earth fraction

The fine earth fraction has similar mineralogy at all depths (Table 3), mainly quartz, a large PDM fraction, kaolinite and gibbsite. The PDM in similar materials has been identified as poorly ordered aluminas (Singh and Gilkes 1995; Tilley and Eggleton 1996), which form by heating. Given that thermal dehydroxilation is the mechanism proposed here for the formation of magnetic concretions, it is possible that some of the PDM is, in fact, poorly ordered alumina. However, the magnetic concretions contain the lowest quantities of PDM, suggesting that the PDM observed in this study is not thermally formed, poorly ordered alumina. We found no XRD or spectroscopic evidence for the presence of poorly ordered alumina, and the nature of the PDM remains unclear.

[FIGURE 2 OMITTED]

Maghemite, hematite, and quartz are the main mineral phases in magnetic nodules (Table 3). Small and intermediate size nodules at the top of the profile have the highest proportions of hematite and maghemite, no PDM, and minor amounts of hydrated mineral phases (cluster A; Table 3). Hematite from magnetic nodules has unusually high calculated Al-substitution values (0.17-3.33 A1/(Fe+A1) mol [mol.sup.-1]). Calculated Al-substitution of hematite in the size fractions 2-4 and 4-8 mm is similar at all depths (<0.20 mol [mol.sup.-1]), but is substantially higher in hematites from the 8-16mm fraction of the A2, B1, and upper B2 horizons (Fig. 7c).

Non-magnetic Fe nodules do not contain maghemite and are grouped into three broad types (clusters D, E, and G; Table 3). Nodules from the A1 horizon have similar mineralogy to nodules and sandstone clasts from the lower parts of the profile (lower B2/CB). They contain broadly equal quantities of quartz, kaolinite, and goethite, and lesser amounts of hematite (samples are grouped in cluster D).

Iron nodules in the A2 and BI horizons (cluster E) have the lowest kaolinite/gibbsite ratios measured during the study, but a high hematite/goethite ratio (Fig. 7e). Goethite is the main mineral in nodules from the upper B2 horizon (cluster G). These samples have the lowest hematite/goethite ratio, and a low kaolinite/gibbsite ratio (0.6 0.7; Fig. 7e, f). Smaller nodules of lower B2 and CB horizons are also grouped in cluster G, but have higher amounts of kaolinite.

Calculated A1 substitution in goethite ranges between 0.15 and 0.27 AI/(Fe+Al) mol [mol.sup.-1] (Fig. 7a). Goethite from A2, B1, and upper B2 horizons is most aluminous, whereas goethites from Fe-rich sandstones in the C horizon are least Al-substituted (0.15-0.17 mol [mol.sup.-1]). Calculated Al substitution in hematites from non-magnetic nodules is mainly between 0.08 and 0.17 Al/(Fe + A1) mol [mol.sup.-1]. Hematites of A2 and B I nodules are significantly more Al-substituted (0.22-0.28 mol [mol.sup.-1]). This coincides with the highest hematite/goethite ratio of the whole profile, but these anomalously high calculated Al-substitution values should be interpreted with caution (see Discussion for more complete treatment of this issue).

[FIGURE 3 OMITTED]

Quartz is the single most important phase in the sandstone clasts of the C and CB horizon (cluster F). They have broadly equal amounts of kaolinite, hematite, and goethite (15 25%), a high hematite/goethite ratio, and the smallest PDM component of the non-magnetic samples. These are the least weathered materials in the profile.

Micromorphology

Grain-supported fabric, in which quartz grains are densely packed and the pore space is partially or completely filled by cement, is most common in Fe nodules and Fe sandstone clasts. The cement is composed of a mixture of Fe and Al oxyhydroxides (hematite, goethite, maghemite, gibbsite) and clay minerals (kaolinite). In Fe nodules, matrix-supported fabric is restricted to the outer cortex, which is of lower porosity and contains less quartz (Fig. 8d).

The cortex is usually a composite of multiple thin lamellae of fine-grained cements, occasionally separated by thin, quartz-rich lenses (Fig. 8a). In concentric nodules in which the core is not spherical, lens-shaped accretions of material usually result in a spherical external morphology. A distinct cortex is common to most non-magnetic and porous magnetic nodules of smooth morphology, but is not evident in the massive magnetic nodules (Fig. 8e). The cortex is also not well developed in nodules of rough external morphology or ferruginous sandstones, which are characterised by protruding quartz grains.

Raman spectroscopy of selected samples confirms that the fine material in the red bands and nodule cores consists mainly of hematite, whereas brown to yellow bands are mainly goethite, with minor amounts of kaolinite and gibbsite. A gradual progression from Fe sandstone clasts to hematitic concentric nodules with a goethitic cortex and then to goethitic uniform/ gradational nodules can be observed. Iron-rich sandstone clasts are typically cemented by kaolinite, goethite, and hematite. Zones rich in goethite are generally restricted to fractures and cracks in the rock (Fig. 8J), while hematite is the main cement in the remainder of the clast. Porosity is low, and etched quartz grains are rare. Concentric nodules with only two distinct morphological zones, such as those common in the A I horizon, are similar to Fe sandstone clasts but have features indicating alteration of hematite to goethite (Fig. 8b). The core of these nodules is most commonly hematitic with a well-developed, gocthitic cortex. Quartz grains are only lightly etched, and hematite is frequently preserved as infillings in these etched grains, even within the goethitic cortex. Goethite coating the pores of the hematitic core suggests that the alteration of hematite to goethite proceeds via dissolution re-precipitation, although preferential removal of hematite in a mixed hematite/ goethite cement is also possible (Fey 1981). A typical example of a small, uniform nodule is shown in Fig. 8c. The matrix material of the nodule is mainly goethite. Hematite infillings are observed in the strongly etched quartz grains which occur throughout the nodule. This is also indicative of the alteration or preferential removal of hematite, but shows that it has proceeded further than in concentric nodules with a goethitic cortex.

[FIGURE 4 OMITTED]

Concentric nodules may also show fcatures typical of the accretion of new layers of material (quartz-free bands, cortexes of multiple fine laminae, quartz-containing lenses/lamellae separating quartz-free ones, lens-shaped accretions; Fig. 8a). This is particularly common in concentric nodules with more than two distinct morphological zones, but may also occur in the less complex concentric nodules. The concentric internal arrangement is usually in the form of alternating rings of red or brown cement with sharp to clear boundaries. In many cases, pores and voids in the nodule core are coated in material similar to that of the cortex.

[FIGURE 5 OMITTED]

Nucleic nodules are similar to concentric nodules and exhibit many of the same features characteristic of accretion. Figure 8d, for example, shows a small, rounded nodule with a well-developed, quartz-poor cortex that is incorporated into the core of a larger nodule. In other nucleic nodules, the nucleus can be of different appearance to the host nodule (e.g. magnetic).

The massive magnetic nodules have a distinctive texture and internal morphology. They are of glassy appearance and very low porosity. Although cemented quartz grains were not evident in the cut nodules, they are readily observed in thin section (Fig. 8e). Cracks filled by an anisotropic, light-coloured material are commonly observed in these nodules and often cut across individual quartz grains.

Discussion

Bourman and co-workers (Milnes et al. 1985, 1987; Bourman 1993a; Bourman 1993b) have built on the work of Hunt et al. (1977) to show that Fe nodules in south-eastern Australia form by contemporary weathering and erosion of Fe-rich sandstone units, ferruginous duricrust, or iron-rich soil mottles and bear only a superficial resemblance to laterite. They conclude that evolution of Fe nodules at such locations occurs through chemical and physical weathering and transport of the preexisting Fe-rich materials. Our results suggest that the Fe nodules in the Poona area are of similar origin.

Iron nodules and weathered sandstone fragments are a widespread component of the surficial soils of the study area (Fig. 2). The greatest concentration of Fe nodules is found in the more steeply sloping, south-western parts of the study area, where Fe-rich sandstones of the Duckinwilla Group crop out and soils are shallow. In marked contrast, Fe nodules are rarely found on the coastal plain, where the sandy soils are generally deeper, topography is subdued, and outcrops are few. Thus, on a catchment scale, there is a clear link between topography, the location of outcropping weathered sandstones, and the occurrence of Fe nodules in surficial soils. It is important to consider the results of our detailed investigation of the ferric soil profile in this context. Furthermore, the profile is at a mid-slope position, immediately below a break in slope. A surface lag of rock fragments and Fe nodules upslope of the profile is evidence of erosion and downslope transport. This suggests that Fe nodules in the upper part of the profile are derived by downslope transport and weathering of iron-rich sandstone clasts similar to those found at the base of the studied profile.

Differentiation of biomantle and ferric horizon

In a soil survey of coastal south-cast Queensland (including the Poona area), Coaldrake (1961) noted that Fe nodules in soils at the base of hills and adjacent to creeks often occur in ferric horizons at depths to 2 m. The frequent concentration of Fe nodules in distinct horizons and the conformity of these horizons with the current land surface led him to believe that the Fe nodules form in situ, as redox accumulations, and that the ferric horizons are the product of seasonal watertable fluctuation. Our evidence shows that it is more likely that the ferric horizons form through downslope transport and differential biological mixing of the soil.

Many insects, arthropods, and small mammals burrow in and through the soil profile. Some particles are too large for the soil fauna to move, and they burrow around such fragments. As tunnels collapse and smaller materials nearby are brought to the surface, these large particles eventually settle downward (Schaetzl and Anderson 2005). In time, they become concentrated at the approximate maximum depth of burrowing. This 'stone line' is overlain by a 'biomantle', which is flee of large particles (Johnson 1990). Breuning-Madsen et al. (2007) have recently demonstrated the role of termite bioturbation in the formation of ferric horizons in West Africa.

[FIGURE 6 OMITTED]

The main agents of bioturbation in humid, subtropical Queensland are ants (Paton et al. 1995). Ant bioturbation is thought to form stone lines consisting of particles larger than ~3 mm in diameter at depths to 2 m, and the homogenisation and mixing of finer soil particles (Schaetzl and Anderson 2005). This is consistent with our observations. Ant mounds are common to the area and the profile morphology strongly suggests that biological mixing of the soil has led to the formation of the ferric B I and B2 horizons and concurrent removal of most nodules from the A2 horizon. The fine-grained biomantle may be derived locally from lower parts of the soil profile, but the ease with which ant mounds are eroded means that it is more likely to contain a substantial upslope component. Apart from the increasing amounts of the gravel-sized nodules to around 1.5m, the upper horizons have a homogeneous, massive structure and contain active and infilled ant tunnels. The mineralogy of the fine earth fraction remains unchanged to a depth of ~1.2m (A2-upper B2 horizons; Table 3). This shows that mixing of the fine earth fraction is efficient to a depth equivalent to the upper B2 horizon. The presence of large quantities of maghemite-rich, magnetic Fe nodules at depths >1 m is further evidence for biological reworking of the soil.

Although various mechanisms have been proposed for its formation (e.g. Taylor and Schwertmann 1974; Barron and Torrent 2002), pedogenic maghemitc is generally thought to form by thermal dehydroxylation of goethite during intense bushfires (Anand and Gilkes 1987). Indeed, thermal formation of maghemite is the only mechanism that has been conclusively shown to occur in the natural soil environment (Grogan et al. 2003). However, conditions conducive to this transformation are restricted to the top 10cm of the soil (Ketterings et al. 2000), strongly implying that the magnetic nodules have been buried since their formation at the soil surface.

[FIGURE 7 OMITTED]

Evolution of the ferric soil profile

Whereas bioturbation can account for the formation of a distinct ferric horizon at a depth of 1-2 m, the distinct mineralogical, morphological, and micromorphological properties of magnetic and non-magnetic Fe nodules at different depths are evidence of a more complex evolutionary process. This includes multiple additions of colluvial material to the soil profile, chemical weathering of nodules, bush-fire-induced thermal alteration of the nodules, as well as in-situ accretion of new generations of Fe/A1 oxihydroxide cement.

The A1 horizon, a recent colluvial addition

Our results show that the A1 horizon is a recent colluvial addition to the profile and has not yet undergone significant mixing with the underlying soil. Iron nodules (magnetic and non-magnetic) comprise 25% of the mass of the A1 horizon, which has a platy texture and a sharp boundary to the underlying, largely nodule-free biomantle. A small number of unaltered, ferruginous sandstone fragments are also present (Fig. 6). Viewed together with the similar overall mineralogy (Table 3), goethite A1 substitution (Fig. 7a), and hematite/ goethite and kaolinite/gibbsite ratios (Fig. 7e, f) of CB horizon ferruginous sandstone clasts and A1 horizon nonmagnetic Fe nodules, this suggests a genetic link between ferruginous sandstone and Fc nodules. It is likely that the nonmagnetic nodules of the A1 horizon are derived from Fe sandstone which has been eroded from locations upslope of the profile and has undergone minor weathering, sufficient only for the conversion of the ferruginous sandstone to Fe nodules, but not sufficient to permit development of the rounded morphology and smooth outer cortex which are characteristics of nodules in the deeper horizons. This interpretation is further supported by the internal morphology and mineralogy of the nodules, which is dominated by concentric (>8mm) and uniform/gradational morphotypes (<8mm). The concentric nodules typically contain only two regions, a hematite-rich red core and a well-developed, yellow, goethite-rich outer band. Such distinct hematitic, red cores and yellow, goethitic cortexes are common features of Fe nodules (Amouric et al. 1986; Herbillon and Nahon 1988; Bourman 1993a). Hematite preserved in cracks of etched quartz grains in the goethite cortex (Fig. 8b) and goethite coatings on pores in the hematite-rich core of some nodules suggest that the distinct zonation is driven by the formation of goethite at the expense of hematite (Fitzpatrick 1988) or by preferential removal of the more strongly pigmenting and more soluble hematite in mixed goethite/hematite cements (Fey 1981; Cornell and Schwertmann 2003). Thus, the concentric appearance of larger nodules in this soil horizon is the result of moderate, centripetal chemical weathering of hematite-rich sandstones and not a result of accretion and growth. Only in the smaller (uniform/gradational) nodules has the destruction of hematite progressed further (Fig. 8c); these are cemented mainly by goethite (Fig. 6). The high hematite/ goethite and kaolinite/gibbsite ratios of A1 nodules compared with upper B2 horizon nodules are also indicative of limited chemical weathering, further evidence that the A1 horizon material is a recent colluvial addition to the soil profile.

[FIGURE 8 OMITTED]

Strongly weathered nodules of the upper ferric horizon

The low kaolinite/gibbsite ratio of the non-magnetic nodules in the A2, B1, and upper B2 horizons suggests that these nodules have been strongly weathered. Intense chemical weathering can lead to desilication of kaolinite and subsequent formation of gibbsite (Huang et al. 2002). Features such as marked quartz etching, replacement of hematite by goethite, and high A1 substitution of goethite and hematite in these nodules are consistent with strong chemical weathering (Schwertmann and Carlson 1994).

Etching of quartz grains is a further indicator of strong chemical weathering. The solubility of quartz is substantially higher in the presence of dissolved Fe (Morris and Fletcher 1987), and repeated redox reactions involving Fe may cause accelerated dissolution of quartz. This process produces very distinctive etch patterns and, in extreme cases, the complete dissolution of quartz grains (Widdowson 2008). In the present study, etched quartz grains were commonly observed in thin section, often filled by hematite or goethite (Fig. 8). Etching is particularly marked in the nodules from A2, B 1, and upper B2 horizons. However, etched quartz grains are rarely observed in ferruginous sandstone clasts (Fig. 8f) and are poorly developed in the Fe nodules of the A1 horizon.

Low kaolinite/gibbsite ratios and high goethite AI substitution are indicative of conditions of high Al activity in the A2 to upper B2 horizon (Fig. 7). The formation and stability of goethite is favoured over that of hematite under conditions of high A1 activity (Fey 1981; Tardy and Nahon 1985). Indeed, nodules from the upper B2 horizon have the lowest hematite/ goethite ratios encountered in this study, so that the conversion of hematite to goethite or selective removal of hematite is near complete. The multiple generations of coatings in pores and the fine lamellae composing the smooth cortex of these nodules are ample evidence of frequent episodes of Fe-oxide dissolution and re-precipitation required for this conversion or removal. The anomalously high hematite/goethite ratios of non-magnetic nodules from A2/B1 horizons contrasts with the general trend, but can be attributed to thermal dehydroxilation of goethite in these non-magnetic nodules (Wells et al. 1989), an interpretation that is supported by the high proportion of magnetic nodules found in A2/B1 horizons.

Accretion of additional material is another important alteration process for the A2, B1, and upper B2 nodules. Of note is the high proportion of concentric and nucleic nodules among the larger nodules of the upper B2 horizon (Fig. 6). The multiple growth or alteration bands of these nodules (up to five distinct zones) and the fact that a large proportion have overgrown well-formed, spherical Fe nodules (Fig. 8d) is ample evidence of their polygenetic nature (Milnes et al. 1987; Bourman 1993a), i.e. of new material accreting in different environments or under different environmental conditions. The internal structure and micromorphology of the nodules suggests that both of these processes are important, i.e. (a) accretion of layers with distinct mineralogical properties in different environments, and (b) alteration of existing materials as a result of changing environmental conditions.

In-situ growth of lower B2 horizon nodules

The lower B2 horizon also contains a large proportion of multi-zone concentric and nucleic nodules, particularly in the >8 mm size fractions (Fig. 6). There are, however, some marked differences between the upper and lower B2 horizon. Bioturbation of the lower B2 horizon and mixing with horizons in the upper profile is absent, as demonstrated by the distinct mineralogy of the fine earth fraction (Table 3) and the almost complete absence of magnetic nodules (Fig. 4) in the lower B2 horizon. Furthermore, the higher kaolinite/gibbsite and hematite/goethite ratios, and lower A1 substitution of both goethite and hematite, indicate that the lower B2 nodules are relatively less chemically weathered (Fig. 7). In terms of mineralogy, the larger nodules in this horizon are, in fact, very similar to those of the A1 horizon and the Fe sandstone of the CB horizon (Table 3), and it is likely that they have experienced limited surface weathering.

The external morphology, internal structure, and micromorphology of the lower B2 nodules indicate extensive in-situ modification. In addition to the high proportion of nucleic and concentric morphotypes, the lower B2 horizon is characterised by a sharp increase in subrounded to angular nodules, as well as those that lack a smooth outer cortex (i.e. protruding quartz grains). These features are caused by the accretion of soil materials cemented by Fe oxides and kaolinite, lending the nodule a rougher, less rounded appearance. Such nodule growth requires an influx of dissolved Fe, most likely derived from reductive dissolution of Fe oxides in underlying horizons during periods of seasonal waterlogging and associated anaerobic conditions (Lohr and Cox 2012). Dissolved ferrous iron can be transported up-profile until it encounters an oxic zone and is immobilised. Horizontal inputs of Fe in groundwater from further upslope are also possible (Bourman 1993a). The accretion of additional layers of cementing Fe oxides is reflected in an increased nodule size in this horizon compared with others (Fig. 4).

Unusually high A1 substitution in magnetic nodules

The unusually high values of A1 substitution obtained for hematite from magnetic nodules in this study must be interpreted with caution. Such high values are not commonly reported for soil hematite (except by Yilley and Eggleton 1996) and shifts in the unit cell parameters of hematite could occur from factors other than A1 substitution of the mineral (e.g. substitution by Cr, Mn, Ni, Cu; Cornell and Schwertmann 2003). However, the trends in hematite A1 substitution observed throughout the profile are entirely consistent with the other data suggesting high-temperature transformation of non-magnetic nodules in the upper parts of the profile. The AI substitution of hematite from magnetic nodules is comparable to the highly aluminous hematites formed by high-temperature dehydroxylation of aluminous goethites under laboratory conditions (Wells et al. 1989; Ruan and Gilkes 1995). Since hematite from non-magnetic nodules in the lower parts of the profile, and therefore not subject to thermal dehydroxylation, has Al substitution within the normal range (0.08-0.17 A1/(Fe+A1) mol [mol.sup.-1]; Cornell and Schwertmann 2003), this suggests that the values of A1-substitution calculated for hematite in this study are valid. That higher A1 substitution in hematites has not been previously reported may simply be a result of sample choice; few studies have investigated the impact of thermal dehydroxylation on natural Fe oxides from soils rich in A1.

A new model for Fe nodule genesis in coastal eastern Australia

This study shows that the Fe nodules found in Poona catchment soils have formed by contemporaneous weathering of Fe-rich sandstone units of the Duckinwilla Group, and erosion and transport of these materials to lower lying areas. Mechanical breakdown and abrasion occurs during transport and bioturbation-controlled burial. There is no evidence to suggest that they are laterite residues, and unlike models invoking redox-dominated formation of Fe nodules, our model does not require an input of dissolved Fe; the sandstone precursor is already cemented by Fc and Al oxyhydroxides. Our findings suggest that the Fe nodule forming processes identified by Hunt et al. (1977) and Bourman and co-workers (Milnes et al. 1985, 1987; Bourman 1993a; Bourman 1993b) in southern Australia apply more broadly to the higher relief, erosive terrains of the eastern Australian coast. What these studies did not consider is the importance of bioturbation in forming the ferric horizons or 'stone lines'. Adding this mechanism to the Hunt/Bourman model explains those features, which led Coaldrake (1961) to believe that the Fe nodules in soils of coastal south-east Queensland were formed wholly in situ.

Summary and conclusions

Iron nodules and weathered sandstone fragments are a widespread component of the surficial soils and sediments of the Fraser Coast. The greatest concentration of Fe nodules is found in the higher elevation, steeply sloping south-western parts of the study area, where Fe-rich sandstones of the Duckinwilla Group crop out and soils are often shallow. The Fe nodules are derived from physical and chemical weathering of these ferruginous sandstones, which are composed mainly of quartz grains cemented by hematite, goethite, and kaolinite. Where these sandstones crop out, they are broken down and transported downslope. Transport and physical/chemical weathering gradually transform these eroded rock fragments into non-magnetic Fe nodules, such as those currently found in the A1 horizon (Fig. 9a).

[FIGURE 9 OMITTED]

The material in the A1 horizon is a recent colluvial addition to the profile and has been only moderately weathered (Fig. 9d). Limited weathering results in nodules with moderate rounding and a rough external morphology. Alteration includes breakdown of the interlocking, cemented grain fabric of the ferruginous sandstones, an increase in porosity with incipient quartz etching, preferential loss of hematite, and conversion of hematite to goethite. This process results in characteristic red, hematitic nodules with a yellow, goethitic cortex. Where the cortex is thick, nodules may have a concentric (two-layer) appearance; where it is thinner, or penetrates the whole nodule, gradational or uniform morphotypes develop.

Nodules in the lower horizons have been buried to form a largely nodule-free 'biomantle' and a ferruginous nodule 'stone line', most likely by ant-induced bioturbation (Fig. 9c). These nodules are older than those in the A1 horizon and are more strongly weathered. This is reflected in the desilication of kaolinite to form gibbsite (A2, B1, and upper B2) and almost complete destruction of hematite to form aluminous goethite (upper B2 horizon). Constant reworking and prolonged weathering results in well-rounded nodules with a smooth outer morphology (upper B2 horizon). The magnetic nodules form by thermal alteration of non-magnetic nodules at the soil surface and are subsequently buried by bioturbation (a more detailed investigation of the genesis and significance of magnetic Fe nodules will be reported elsewhere).

Reductive dissolution of Fe oxides in underlying horizons during periods of seasonal waterlogging results in translocation of ferrous iron up-profile, where it precipitates as Fe-oxide overgrowths on nodules of the B2 horizon. This accretion of new material results in nodule growth and more angular, rough external morphology that is preserved because of a lack of disruption by bioturbation. Multiple-zone nucleic and concentric morphotypes are common in this horizon and are likely to have experienced several episodes of accretion under a range of environmental conditions. Relatively unaltered, Fe-rich sandstone clasts were found at the base of the profile and are equivalent to the upslope materials from which the Fe nodules were originally derived.

doi. 10.1071/SR12372

Received 20 December 2012, accepted 12 March 2013, published online 10 April 2013

Acknowledgments

This work was supported by the Australian Research Council (Linkage Project LP0669786), Forest Plantations Queensland, and the Institute for Sustainable Resources at Queensland University of Technology. Grateful acknowledgements are made to Nicole Harb for assistance in the field and laboratory, to Luke Nothdurft for advice on sample preparation and to Llew Rintoul for help with the Raman spectroscopy analyses.

References

Amouric M, Baronnct A, Nahon D, Didier P (1986) Electron microscopic investigations of iron oxyhydroxides and accompanying phases in lateritic iron-crust pisolites. Clays and Clay Minerals 34, 45-52. doi:10.1346/CCMN.1986.0340106

Anand R (2001) Evolution, classification and use of ferruginous regolith materials in gold exploration, Yilgarn Craton, Western Australia. Geochemistry Exploration Environment Analysis I, 221-236. doi:10.1144/geochem.1.3.221

Anand RR, Gilkes RJ (1987) The association of maghemite and corundum in Darling Range laterites, Western Australia. Australian Journal of Soil Research 25, 303-311. doi:10.1071/SR9870303

Anand RR, Paine M (2002) Regolith geology of the Yilgam Craton, Western Australia: implications for exploration. Australian Journal of Earth Sciences 49, 3-162. doi:10.1046/j.1440-0952.2002.00912.x

Barron V, Torrent J (2002) Evidence for a simple pathway to maghemite in Earth and Mars soils. Geochimica et Cosmochimica Acta 66, 2801-2806. doi:10.1016/S0016-7037(02)00876-1

Blume HP (1968) Zum Mechanismus der Marmorierung und Konkretionsbildung in Stauwasserbrden. Zeitschrift fur Pflanzenernahrung und Bodenkunde 119, 124-134. doi:10.1002/jpln.19681190205

Bourman RP (1993a) Modes of ferricrete genesis evidence from southeastern Australia. Zeitschrift fur Geomorphologie 37, 77-101.

Bourman RP (1993b) Perennial problems in the study of laterite-a review. Australian Journal of Earth Sciences 40, 387-401. doi:10.1080/08120099308728090

Breuning-Madsen H, Awadzi TW, Koch CB, Borggaard OK (2007) Characteristics and genesis of pisolitic soil layers in a tropical moist semi-deciduous forest of Ghana. Geoderma 141, 130 138. doi: 10.1016/j.geoderma.2007.05.009

Childs CW (1975) Composition of iron-manganese concretions from some New Zealand soils. Geoderma 13, 141-152. doi:10.1016/0016-7061(75) 90063-4

Coaldrake JE (1961) 'The ecoystem of the coastal lowlands ("Wallum") of southern Queensland." (CSIRO: Melbourne)

Cornell RM, Schwertmann U (2003) "The iron oxides. Structure, properties, reactions, occurences and uses.' (Wiley-VCH: Weinheim, Germany)

Coventry R J, Taylor RM, Fitzpatrick RW (1983) Pedological significance of the gravels in some red and grey earths of central North Queensland. Australian Journal of Soil Research 21, 219-240. doi:10.1071/SR9830219

Cranfield LC (1994) '1:250000 Geological Series--Explanatory Notes. Maryborough, Queensland. Sheet SG56-6.' (Geological Survey of Queensland, Department of Minerals and Energy: Brisbane, Qld)

Faniran A (1970) Maghemite in the Sydney duricrusts. The American Mineralogist 55, 925-933.

Fey MV (1981) Hypothesis for the pedogenic yellowing of red soil materials. In "Proceedings 10th National Congress Soil Science Society Southern Africa'. Technical Communication 180, Department of Agriculture, Pretoria. pp. 130-136. (IUSS)

Fitzpatrick RW (1988) Iron compounds as indicators of pedogenic processes: Examples from the southern hemisphere. In 'Iron in soils and clay minerals'. (Eds JW Stucki, BA Goodman, U Schwertmann) pp. 351-396. (D. Reidel Publishing Company: Dordrecht, The Netherlands)

Gasparatos D, Haidouti C, Tarcnidis D (2004) Characterization of iron oxides in Fe-rich concretions from an imperfectly-drained Greek soil: a study by selective-dissolution techniques and X-ray diffraction. Archives of Agronomy and Soil Science 50, 485-493. doi:10.1080/0365034042000216149

Grogan KL, Gilkes RJ, Lottermoser BG (2003) Maghemite formation in burnt plant litter at East Trinity, North Queensland, Australia. Clays and Clay Minerals 51, 390-396. doi:10.1346/CCMN.2003.0510404

Hanesch M (2009) Raman spectroscopy of iron oxides and (oxy) hydroxides at low laser power and possible applications in environmental magnetic studies. Geophysical Journal International 177, 941-948. doi:10.1111/j.1365-246X.2009.04122.x

Herbillon A J, Nahon D (1988) Laterites and laterization processes. In 'Iron in soil and clay minerals'. (Eds JW Stucki, BA Goodman, U Schwertmann) pp. 779-796. (D. Riedel: Dordrecht, The Netherlands)

Huang PM, Wang MK, Kampf N, Schulze DG (2002) Aluminium hydroxides. In 'Soil mineralogy with environmental applications'. (Eds JB Dixon, DG Schulze) pp. 261-289. (Soil Science Society of America: Madison, WI)

Hunt PA, Mitchell PB, Paton TR (1977) "Laterite profiles' and 'lateritic ironstones' on the Hawkesbury Sandstone, Australia. Geoderma 19, 105 121. doi:10.1016/0016-7061(77)90019-2

Isbell RF (2002) 'The Australian Soil Classification.' (CSIRO Publishing: Melboume)

IUSS Working Group WRB (2006) 'World reference base for soil resources 2006.' World Soil Resources Report No. 103. (FAO: Rome)

Johnson DL (1990) Biomantle evolution and the redistribution of earth materials and artifacts. Soil Science 149, 84-102. doi:101097/00010694-199002000-00004

Ketterings QM, Bigham JM, Laperche V (2000) Changes in soil mineralogy and texture caused by slash-and-burn fires in Sumatra, Indonesia. Soil Science Society of America Journal 64, 1108-1117. doi:10.2136/sssaj2000.6431108x

Latrille C, Elsass F, van Oort F, Denaix L (2001) Physical speciation of trace metals in Fe-Mn concretions from a rendzic lithosol developed on Sinemurian limestones (France). Geoderma 100, 127-146. doi:10.1016/S0016-7061(00)00083-5

Lohr SC, Cox ME (2012) The influence of vegetation and soil type on the speciation of iron in soil water. European Journal of Soil Science 63, 377 388. doi:10.1111/j.1365-2389.2012.01435.x

Milnes AR, Bourman RP, Northcote KH (1985) Field relationships of ferricretes and weathered zones in southern South Australia: a contribution to 'laterite' studies in Australia. Australian Journal of Soil Research 23, 441-465. doi: 10.1071/SR9850441

Milnes AR, Bourman RP, Fitzpatrick JW (1987) Petrology and mineralogy of 'laterites' in southern and eastern Australia and southern Africa. Chemical Geology 60, 237 250. doi:10.1016/0009-2541(87)90129-X Morris BA, Fletcher IA (1987) Increased solubility of quartz following ferrous-ferric iron reactions. Nature 330, 558-561. doi:10.1038/330558a0

Paton TR, Humphreys GS, Mitchell PB (1995) 'Soils. A new global view.' (Routledge: London)

Ruan H, Gilkes R (19951 Dehydroxylation of aluminous goethite: Unit cell dimensions, crystal size and surface area. (Tars and Clay Minerals 43, 196-211. doi:10.1346/CCMN.1995.0430207

Schaetzl R, Anderson S (2005) 'Soils--genesis and geomorphology.' (Cambridge University Press: Cambridge, UK)

Scheinost AC, Schwertmann U (1999) Color identification of iron oxides and hydroxysulfates--use and limitations. Soil Science Society of America Journal 63, 1463-1471.

Schulz MS, Vivit D, Scbulz C, Fitzpatrick J, White A (2010) Biologic origin of iron nodules in a marine terrace chronosequence, Santa Cruz, California. Soil Science Society of America Journal 74, 550-564. doi:10.2136/sssaj2009.0144

Schwertmann U, Carlson L (1994) Aluminum influence on iron oxides: XVII. Unit-cell parameters and aluminum substitution of natural goethites. Soil Science Society of America Journal 58, 256-261. doi:l0.2136/sssaj1994.03615995005800010039x

Schwertmann U, Fanning DS (1976) Iron-manganese concretions in hydrosequenccs of soils in loess in Bavaria. Soil Science Society of America Journal 40, 731-738. doi:10.2136/sssaj1976.03615995004000050034x

Schwertmann U, Fechter H (1984) The influence of aluminum on iron oxides: XI. Aluminum-substituted maghemitc in soils and its formation. Soil Science Society of America Journal 48, 1462-1463. doi:l0.2136/sssaj1984.03615995004800060054x

Schwertmann U, Friedl J, Stanjek H, Schulze DG (2000) The effect of A1 on Fe oxides. XIX. Formation of Al-substituted hematite from ferrihydritc at 25[degrees]C and pH 4 to 7. Clays and Clay Minerals 48, 159-172. doi:10.1346/CCMN.2000.0480202

Singh B, Gilkes RJ (1995)The natural occurrence of chi-alumina in lateritic pisolites. Clay Minerals 30, 39-44. doi: 10.1180/claymin.1995.030.1.04

Singh B, Gilkes RJ (1996) Nature and properties of iron rich glaebules and mottles from some south-west Australian soils. Geoderma 71, 95-120. doi: 10.1016/0016-7061(95)00092-5

Tardy Y (1992) Diversity and terminology of lateritic profiles. In 'Weathering, soils and paleosols'. (Eds IP Martini, W Chesworth) pp. 379-405. (Elsevier: Amsterdam)

Tardy Y (1997) 'Petrology of laterites and tropical soils.' (A. A. Balkema: Rotterdam)

Tardy Y, Nahon D (1985) Geochemistry of laterites, stability of Al-goethite, A1-hematite and Fe-kaolinite in bauxites and ferricretes: an approach to the mechanism of concretion formation. American Journal of Science 285, 865-903. doi:l0.2475/ajs.285.10.865

Taylor G, Eggleton RA (2008) Genesis of pisoliths and of the Weipa Bauxite deposit, northern Australia. Australian Journal of Earth Sciences 55, S87-S103. doi:10.1080/08120090802438274

Taylor RM, Schwertmann U (1974) Maghemite in soils and its origin. 11. Maghemite syntheses at ambient temperature and pH 7. Clay Minerals 10, 299-310. doi:10.1180/claymin.1974.010.4.08

Tilley D, Eggleton RA (1996) The natural occurrence of eta-alumina (eta[Al.sub.2][O.sub.3]) in bauxite. Clays and Clay Minerals 44, 658-664. doi:10.1346/CCMN. 1996.0440508

Tripathi JK, Rajamani V (2007) Geochemistry and origin of ferruginous nodules in weathered granodioritic gneisses, Mysore Plateau, Southern India. Geochimica et Cosmochimica Acta 71, 1674-1688. doi:10.1016/j.gca.2007.01.001

Wells MA, Gilkes RJ, Anand RR (1989) The formation of corundum and aluminous hematite by the thermal dehydroxylation of aluminous goethite. Clay Minerals 24, 513-530. doi:10.1180/claymin.1989.024.3.05

White GN, Dixon JB (1996) Iron and manganese distribution in nodules from a young Texas Vertisol. Soil Science Society of America Journal 60, 1254-1262. doi:10.2136/sssaj1996.03615995006000040042x

Widdowson M (2008) Laterite and Ferrierete. In 'Geochemical sediments and landscapes'. (Eds DJ Nash, SJ McLaren) pp. 46 94. (Wiley-Blackwell: Chichester, UK)

Zhang M, Karathanasis AD (1997) Characterization of iron-manganese concretions in Kentucky alfisols with perched water tables. Clays and Clay Minerals 45, 428-139. doi:10.1346/CCMN.1997.0450312

S. C. Lohr (A,B,E), M. Grigorescu (A,C), and M. E. Cox (A,D)

(A) School of Earth, Environment and Biological Sciences, Science and Engineering Faculty, Queensland University of Technology, Brisbane, Qld 4001, Australia.

(B) School of Earth and Environmental Sciences, University of Adelaide, Adelaide, SA 5005, Australia.

(C) Geological Survey of Queensland.

(D) National Centre for Groundwater Research and Training.

(E) Corresponding author. Email: stefan.loehr@adelaide.edu.au
Table 1. Field description of ferric soil profile
in Poona catchment

Terminology after Australian Soil and Land Survey Field Handbook
(National Committee on Soil and Terrain 2008).
Location: 0485856 E, 7145385 N

Depth
(cm)      Horizon     Colour     Description

0-15        A1      2.5YR 4/6    Clay loam, sandy. Platy structure.
                                 Fine roots, faunal channels,
                                 bioturbation. Some nodules on soil
                                 surface, but mainly concentrated
                                 at bottom of horizon. 30% nodules,
                                 matrix-supported, 0.5-2 cm
                                 diameter. Subrounded-rounded.
                                 Sharp, smooth boundary

15-55       A2      2.5YR 4/6    Sandy clay loam. Massive, very
                                 firm. Fine-thick roots, ant
                                 tunnels, krotovinas, evidence of
                                 bioturbation. Few nodules, <5%,
                                 0.2-1.2 cm. Rounded, high
                                 sphericity. Diffuse boundary

55-75       B1      2.5YR 4/8    Clay loam. Ant tunnels, krotovinas.
                                 Nodules 10% with gradual increase
                                 to bottom, 0.5-2 cm. Rounded,
                                 medium-high sphericity. Clear,
                                 wavy boundary

75-175      B2      7.5YR 4/6    Clay loam. Ferric horizon. Nodules
                                 80-90%, clast-supported. Decrease
                                 from rounded well rounded at top,
                                 to angular-subrounded. Mostly
                                 medium-high sphericity. Clear,
                                 irregular boundary

175-210     CB       10YR 5/6    Clay loam, sandy. Massive
                                 structure. Up to 50% poorly sorted,
                                 ferruginised rock clasts, 4-8 cm
                                 in size. Subangular angular

>210         C       10YR 5/6    As above, smaller and less
                                 frequent rock fragments. (Sandy)
                                 light-medium clay

Table 2. Characteristics of Fe nodule morphotypes

Morphotype     Properties

Uniform/       Non-magnetic
gradational    Most commonly observed type <8 mm
               Broadly uniform internal morphology [or.bar] gradual
                 differentiation between zones-porosity, colour
                 change, or grain size
               Quartz grains cemented by Fe/Al oxides
               Medium-high porosity
Concentric     Non-magnetic
               Complex internal morphology
               Two or more sharply defined zones--distinct cement
                 colour, porosity, grain-size, or cement to
                 quartz ratio
               Broadly concentric arrangement of zones
               Greater number of zones in larger nodules
               Most common in the >8 mm size fraction, but also
                 found in smaller fraction
Nucleic/       Non-magnetic
compound       Most complex morphotype
               Contain one or more cemented nodule, Fe-rock, or
                 nodule fragment
               Nuclei may be of different morphology or even magnetic
               Similar to gradational or concentric nodules
               Most common in >8 mm fractions, but also found in
                 smaller fractions
Porous         More porous, less strongly magnetic than massive
magnetic         magnetic
               Rusty red to very dark red/black
               Most commonly porous core and harder, low-porosity
                 outer rind
               Colour may vary between the core and the outer rind
               Similar structure to gradational/uniform non-magnetic
                 nodules
Massive        Most common magnetic morphotype
magnetic       Mainly <8 mm
               Uniform internal morphology
               Strongly magnetic
               Extremely hard, difficult to break open
               Distinctive conchoidal fracture
               Very dark red to black
               Shiny, vitreous appearance
               Very low porosity
               No/poorly expressed rind
               Cemented quartz grains difficult to observe
               Occasional large cracks filled with lighter material

Table 3. Mineralogy (wt%) of Fe nodules

PDM, Poorly diffracting materials; M, magnetic; NM, non-magnetic; FR,
sandstone clasts; E, fine earth fraction. X-Ray diffractograms were
clustered using whole-pattern hierarchical clustering; samples grouped
in a cluster have similar mineralogy, e.g. the fine earth fraction
(<2 mm) is dominated by quartz, PDM, and kaolinite and is grouped in
cluster X; there is a clear separation between fine earth fraction,
magnetic and non-magnetic Fe nodules; mineralogical changes with depth
can also be seen in magnetic and non-magnetic nodules. Note that
although magnetic Fe nodules were observed in lower soil horizons,
quantities were very low (<0.2 wt%); magnetic Fe nodules are not
characteristic of the subsoil

Depth (cm)    Size (mm)    Type     Cluster     Quartz

0-15             <2         E          X          50
                 24         M          A          17
                            NM         D          27
                 48         M          B          23
                            NM         D          24
                8-16        M          C          28
                            NM         D          25
                 >16        NM         D          28

15-55            <2         E          X          50
                 2-4        M          A          16
                            NM         E          41
                 4-8        M          A          25
                            NM         E          37
                8-16        M          B          29
                            NM         E          33

55-75            <2         E          X          47
                 2-4        M          A          16
                            NM         E          41
                 4-8        M          B          21
                            NM         E          32
                8-16        M          C          28
                            NM         E          26
                 >16        NM         E          34

85-115           <2         E          X          43
                 2-4        M          B          15
                            NM         G          28
                 4-8        M          C          23
                            NM         G          27
                8-16        M          C          27
                            NM         G          27
                 >16        NM         G          27

140-175          <2         E          X          42
                 2-4        M          B          23
                            NM         G          25
                 48         NM         G          26
                8- 16       NM         D          27
                 >16        NM         D          28

175-210          <2         E          X          51
                 24         FR         G          39
                 48         M          B          22
                            FR         D          32
                8-16        FR         D          31
                 >16        FR         F          34

210--230         <2         E      [X.sub.1]      54
                 2-4        FR     [X.sub.2]      66
                 48         FR         F          36
                8-16        FR         F          34
                 >16        FR         F          33

Depth (cm)    Hematite   Goethite   Maghemite

0-15             2          3           0
                 46         1           34
                 9          20          0
                 35         4           33
                 10         25          0
                 23         5           23
                 10         27          0
                 16         25          0

15-55            1          2           0
                 47         1           35
                 14         10          0
                 35         1           37
                 15         11          0
                 24         3           26
                 15         15          0

55-75            2          3           0
                 44         1           35
                 13         12          0
                 31         3           37
                 14         16          0
                 25         6           20
                 11         25          0
                 16         18          1

85-115           1          3           0
                 43         3           33
                 8          25          1
                 29         6           31
                 6          32          0
                 22         6           25
                 5          33          0
                 6          31          0

140-175          1          5           0
                 32         5           34
                 2          18          0
                 4          25          0
                 10         25          0
                 15         28          0

175-210          1          3           0
                 5          18          0
                 34         5           32
                 9          28          0
                 10         23          0
                 20         21          0

210--230         0          2           0
                 2          5           1
                 11         17          0
                 18         19          0
                 24         16          0

Depth (cm)    Kaolin    Gibbsite    PDM

0-15             9         8        28
                 0         2         1
                19         5        19
                 0         4         0
                20         3        17
                 2         5        13
                17         3        17
                17         2        12

15-55           11         10       25
                 0         1         0
                 8         9        19
                 0         2         0
                10         8        19
                 3         3        12
                14         6        18

55-75           12         12       24
                 1         3         0
                 9         12       11
                 2         5         0
                 9         11       17
                 2         8        12
                13         9        15
                19         12        1

85-115          14         14       25
                 0         5         0
                15         10       15
                 0         7         4
                16         8        11
                 2         7        12
                17         8        10
                16         8        10

140-175         20         10       22
                 0         5         0
                28         4        22
                25         2        18
                20         1        17
                18         1        10

175-210         16         4        25
                18         2        19
                 0         5         3
                18         1        12
                18         1        17
                15         0        10

210--230        18         2        24
                 9         0        17
                21         1        14
                19         1        10
                15         0        11
COPYRIGHT 2013 CSIRO Publishing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2013 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Lohr, S.C.; Grigorescu, M.; Cox, M.E.
Publication:Soil Research
Article Type:Report
Geographic Code:8AUST
Date:Mar 1, 2013
Words:9517
Previous Article:Salinity affects the response of soil microbial activity and biomass to addition of carbon and nitrogen.
Next Article:Ensemble pedotransfer functions to derive hydraulic properties for New Zealand soils.
Topics:

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters