Duricrust Formation

Cap structures within silcretes have long been used as a diagnostic indicator of pedogenic silicification. However, a growing number of studies of the micromorphology of non-pedogenic silcretes indicate that this may no longer be appropriate. This paper presents the first systematic investigation of the micro-fabric, geochemistry and mineralogy of cap structures in groundwater silcretes, through an analysis of conglomeratic varieties (puddingstones) from the southern UK. Our results suggest that cap structures in groundwater silcretes fall within a spectrum of types, related to the degree of sorting in the inter-gravel host sediment. At one end of this spectrum are well-defined caps within otherwise well-sorted, overgrowth-dominated silcretes. These caps exhibit a grain-supported fabric, are cemented by micro-and/or cryptocrystalline silica, and contain floating silt-sized quartz and Ti-oxide grains. We propose that these structures developed mainly as a result of in-washing of fine sediments that were subsequently silicified. At the other end of the spectrum are silcretes with caps defined by concentrations of Ti-oxide grains, as opposed to cement type and grain size. These formed mainly as a result of the remobilisation and precipitation of Ti during the silicification of gravels containing interstitial clay-rich sandy sediment. Between these end-members are silcretes with cap structures formed by a combination of in-washing and redistribution of fines plus some local remobilisation of Ti. Overall, the cap structures in this study exhibit a simple micromorphology, lacking the alternating Ti-and silica-rich lamellae typical of pedogenic silcrete. We conclude that the presence of cap structures alone should not be considered diagnostic of pedogenic silicification unless accompanied by other indicators such as a differentiated profile and abundant, complex, way-up structures within the micro-fabric.

Dryland alluvial fans developed along the northern flanks of the Sierra Lisbona in the north-western Vera basin (Almería region, southeast Spain) display negligible amounts of incision despite distal base-level lowering of N30 m by the modern drainage network. The fans are of early-middle Quaternary age and are graded to a coeval river terrace, now isolated from the modern drainage network. The negligible incision is due to the encasement and 'fossilisation' of the alluvial fans by calcrete. This paper explores the reasons for such fossilisation and assesses the relative importance of pedogenic and groundwater mechanisms for calcrete formation within an alluvial fan setting. A single fan was selected for detailed examination. The geomorphological and sedimentological features of the fan, its catchment area and their relationship to the distal river terrace were documented. Qualitative and semi-quantitative petrographic and scanning electron microscope analyses of calcrete samples collected from transects across the fan surface, and within its distal top river terrace surface, enabled the style, pattern and relative timing of calcrete development to be assessed. Calcrete fabrics comprised initial micritic grain-coating cements, pellets and glaebular carbonate nodules, with interstitial spaces infilled by equant sparite and microsparite mosaics. It is proposed that the early phases of calcrete development were dominated by pedogenic processes with increasing groundwater calcretisation over time. Point count data indicated increased quantities of interstitial sparite and microsparite cement within near-surface proximal fan calcretes and at depth across the fan, suggesting that groundwater processes played a more important role in calcrete formation in these locations. The contribution of groundwater to calcrete development can be best explained by the intrinsic funnelling of groundwater from the catchment through the proximal fan head area, a zone where the fan gravels are thinnest. Calcrete 'fossilisation' appears to have followed a reduction in the fan catchment area as a result of rockfalls and watershed stream capture, which reduced water and sediment supply to the fan and enabled surface stabilization and calcrete development to take place. The reduced sediment/water supply, combined with calcrete fossilisation, appears to have protected the alluvial fans from regional base-level lowering. The implications of these results for existing pedostratigraphic models of calcrete development in alluvial fans are subsequently explored.

Lithic artifacts from the African Middle Stone Age (MSA) offer an avenue to explore a range of human behaviors, including mobility, raw material acquisition, trade and exchange. However, to date, in southern Africa it has not been possible to provenance the locations from which commonly used stone materials were acquired prior to transport to archaeological sites. Here we present results of the first investigation to geochemically fingerprint silcrete, a material widely used for tool manufacture across the subcontinent. The study focuses on the provenancing of silcrete artifacts from the MSA of White Paintings Shelter (WPS), Tsodilo Hills, in the Kalahari Desert of northwest Botswana. Our results suggest that: (i) despite having access to local quartz and quartzite at Tsodilo Hills, MSA peoples chose to transport silcrete over 220 km to WPS from sites south of the Okavango Delta; (ii) these sites were preferred to silcrete sources much closer to Tsodilo Hills; (iii) the same source areas were repeatedly used for silcrete supply throughout the 3 m MSA sequence; (iv) during periods of colder, wetter climate, silcrete may have been sourced from unknown, more distant, sites. Our results offer a new provenancing approach for exploring prehistoric behavior at other sites where silcrete is present in the archaeological record.

Africa and yet have neither been analysed systematically nor in any detail. In this study, the petrological and geochemical characteristics of suites of calcrete, silcrete and intergrade duricrusts from two fresh, relatively deep exposures at Kang Pan and Tswaane (adjacent to the Okwa Valley) in the central Kalahari, Botswana, are described. The duricrust profile at Kang Pan consists of a highly indurated crystalline non-pedogenic calcrete which has been extensively silicified by chalcedony, and, rarely, cryptocrystalline silica or microquartz. Silicification is most extensive in lower parts of the profile, where replacement is related to groundwater fluctuations, and in upper sections due to periodic flooding by ephemeral surface water. The exposure at Tswaane consists of a sequence of pale green glauconitic non-pedogenic silcrete and cal-silcrete overlain by non-pedogenic calcrete, all of which have formed within sediments situated upon granitoid-gneiss bedrock. The siliceous duricrusts are dominated by cryptocrystalline silica cements and appear to have developed through the replacement of a preexisting non-pedogenic calcrete. These siliceous duricrusts have also been calcified at a later date during the formation of the overlying calcrete to produce a complex range of silica-carbonate cements. At both sites, the style and type of silicification present appears to be determined by the duration of wetting and the permeability of the precursor calcrete. Geochemical evidence indicates a lack of chemical weathering profile development within the granitoid-gneiss bedrock and considerable differences between the chemical signature of the bedrock and combined duricrusts from Kang and Tswaane. This suggests that bedrock made a minimal contribution in terms of silica and carbonate species to duricrust formation, and that the majority of cementing agents were non-local. It would therefore appear likely that the geomorphological context of each site had a major influence upon the development of calcrete, silcrete and intergrade duricrust cements. where silcretes, calcretes and intergrade duricrust varieties form a major component of the Kalahari Group sediments (South African Committee for Stratigraphy, 1980). Three main silcrete-calcrete intergrade types have been identified in this region on the basis of silica-carbonate associations within the duricrust cement . These are (a) duricrusts where extensive secondary silicification has occurred within a calcareous matrix; (b) varieties where secondary carbonate has been precipitated within a siliceous matrix; and (c) materials where silica and carbonate matrix cements appear to have been precipitated either contemporaneously or in close succession. Further sub-types have been identified within each of these three groups, dependent upon whether secondary materials have either replaced pre-existing cements or have simply been precipitated in pore spaces and voids within a pre-existing geochemical sediment.

This paper presents the first intensive integrated field and laboratory study of an area of typical silcrete occurrence in the UK, and enables parallels to be drawn with in situ silcretes in the neighbouring Paris Basin and other parts of northern Europe. Silcrete distribution in the eastern South Downs is localized and discontinuous, with occurrence principally as dislocated boulders on the Chalk. These boulders mainly occur between the Cuckmere and Goldstone valleys, with greatest concentrations around Stanmer, Falmer, Rottingdean, and in the Goldstone valley. Many occur at higher levels in the landscape, on or near interfluves usually peripheral to outliers of Palaeocene sediments and in close association with Claywith-Flints. At lower levels they occur on valley floors and on the coastal plain within Quaternary sediments. Boulders commonly have a-axis dimensions of 0·5-2 m and thicknesses of 0·3-0·6 m. Silcretes in higher positions typically exhibit angular tabular or prismatic shapes whilst those in derived settings are more rounded, suggesting weathering and erosion during transport from a localized high level silcrete lens (or lenses). Three varieties of silcrete have been identified. Pale grey 'saccharoid sarsens' are most widespread, and exhibit a simple grain-supported (GS-) fabric with predominantly sand-sized quartzose sediment cemented by optically continuous quartz overgrowths and minor micro-and cryptocrystalline silica. Brown 'hard sarsens' occur as smaller blocks, mainly in the Seaford Head area. These also exhibit a GS-fabric but with more fine sediment and a greater variety of cement types. Textural and geochemical evidence suggests there is gradation between the saccharoid and hard varieties of sarsen. Conglomeratic 'puddingstone' is common in the Goldstone valley but rare elsewhere, and shows closest similarity to saccharoid sarsen in texture and cementation. Silcrete micromorphology is universally simple and uniform, with fabrics suggestive of groundwater or drainage-line silicification. The uniformity of sediments within all three types of silcrete indicates a common host, suggested to be the Upnor Formation of the Lambeth Group (Palaeocene). Deposition of the host sediment appears to have been followed by illuviation of clay-rich material via the primary fabric and along cracks to form geopetal caps, drapes and vein-like structures. During silicification this finer material has been replaced by less well-ordered silica whilst optically continuous quartz overgrowths characterize 'purer' areas. The silcrete was also affected by late-stage influxes of iron oxides, clay minerals and occasionally calcite. On the basis of distributional and other evidence it is suggested that silicification occurred during the Neogene or Quaternary, in association with acid leaching of Lambeth Group sediments probably under temperate conditions. The timing of silcrete formation in the eastern South Downs is thus much later than proposed for other parts of the UK.

A synthesis of the geochemistry of silcretes and their host sediments in the Kalahari Desert and Cape coastal zone, using isocon comparisons, shows that silcretes in the two regions are very different. Kalahari Desert silcretes outcrop along drainage-lines and within pans, and formed by groundwater silicification of near-surface Kalahari Group sands. Silicification was approximately isovolumetric. Few elements were lost; silicon (Si) and potassium (K) were gained as microquartz precipitated in the sediment porosity and glauconite formed in the sub-oxic groundwater conditions. The low titanium (Ti) content reflects the composition of the host sands. Additional elements in the Kalahari Desert silcretes were supplied in river water and derived from weathering of silicates in basement rocks. Evaporation under an arid climate produced high-pH groundwater that mobilized and precipitated Si; this process is still occurring. In the Cape coastal zone, pedogenic silcretes cap hills and plateaus, overlying deeply weathered argillaceous bedrock. Silicification resulted from intensive weathering that destroyed the bedrock silicates, almost completely removing most elements and causing a substantial volume decrease. Some of the silica released formed a microcrystalline quartz matrix, and most Ti precipitated as anatase, so the Cape silcretes contain relatively high Ti levels. The intense weathering that formed the Cape silcretes could have occurred in the Eocene, during and after the Palaeocene-Eocene Thermal Maximum, when more acidic rainfall and high temperatures resulted in intensified silicate weathering worldwide. This could have been responsible for widespread formation of pedogenic silcretes elsewhere in Africa and around the globe. Trace element sourcing of silcrete artefacts to particular outcrops has most potential in the Cape, where differences between separate bedrock areas are reflected in the silcrete composition. In the Kalahari Desert, gains of some elements can override compositional differences of the parent material, and sourcing should be based on elements that show the least change during silicification.

Archaeological sites in northern Africa provide a rich record of increasing importance for the origins of modern human behaviour and for understanding human dispersal out of Africa. However, the timing and nature of Palaeolithic human behaviour and dispersal across northwestern Africa (the Maghreb), and their relationship to local environmental conditions, remain poorly understood. The cave of Rhafas (northeast Morocco) provides valuable chronological information about cultural changes in the Maghreb during the Palaeolithic due to its long stratified archaeological sequence comprising Middle Stone Age (MSA), Later Stone Age (LSA) and Neolithic occupation layers. In this study, we apply optically stimulated luminescence (OSL) dating on sand-sized quartz grains to the cave deposits of Rha-fas, as well as to a recently excavated section on the terrace in front of the cave entrance. We hereby provide a revised chronostratigraphy for the archaeological sequence at the site. We combine these results with geological and sedimentological multi-proxy investigations to gain insights into site formation processes and the palaeoenvironmental record of the region. The older sedimentological units at Rhafas were deposited between 135 ka and 57 ka (MIS 6 –MIS 3) and are associated with the MSA technocomplex. Tanged pieces start to occur in the archaeological layers around 109 ka, which is consistent with previously published chronological data from the Maghreb. A well indurated duricrust indicates favourable climatic conditions for the pedogenic cementation by carbonates of sediment layers at the site after 57 ka. Overlying deposits attributed to the LSA technocomplex yield ages of ~21 ka and ~15 ka, corresponding to the last glacial period, and fall well within the previously established occupation phase in the Maghreb. The last occupation phase at Rhafas took place during the Neolithic and is dated to ~7.8 ka.