Elemental differences: Geochemical identification of aboriginal silcrete sources in the Arcadia Valley, eastern Australia Grant W.G. Cochrane a , John A. Webb b, ⁎, Trudy Doelman c , Phillip J. Habgood d,e a 15/52–58 Meta St, Mooloolaba, Queensland 4557, Australia b Environmental Geoscience, Department of Ecology, Environment and Evolution, La Trobe University, Bundoora, Vic 3086, Australia c Department of Archaeology, University of Sydney, NSW 2006, Australia d ARCHAEO Cultural Heritage Services, PO Box 333, The Gap, Queensland 4061, Australia e School of Social Science, The University of Queensland, St Lucia, Queensland 4072, Australia abstract article info Article history: Received 11 April 2016 Received in revised form 11 November 2016 Accepted 16 November 2016 Available online xxxx Portable X-ray Fluorescence (PXRF) analysis was applied to silcrete artefacts from surface concentrations in the southern Arcadia Valley, eastern Australia. These artefacts were manufactured from river cobbles, as shown by the waterworn cortex, and could have been obtained from three sources: Dawson River, Carnarvon Creek and/ or Clematis Creek. PXRF analyses of cobbles from these sites, evaluated using nonparametric statistics (because the data are skewed) and confidence ellipses, showed that the three sources can be distinguished by their Fe and Zr concentrations. Comparison with artefact analyses showed that many artefacts were probably sourced from Clematis Creek, with a substantial number from Dawson River but few, if any, from Carnarvon Creek. The sourcing pattern indicates that Aboriginal people in the Arcadia Valley were influenced not only by proximity when procuring silcrete, but also quality, cobble size and traditional mobility strategies. Carnarvon Creek was ig- nored because of its distance from the artefact sites and lower quality silcrete. Clematis Creek was the most favoured source because although distant from the artefact sites, it contained large silcrete cobbles (sufficient for the manufacture of large scrapers and knives), and was probably visited frequently due to its central location in the traditional country of the Karingbal People. © 2016 Elsevier Ltd. All rights reserved. Keywords: Silcrete Artefact sourcing Portable X-ray fluorescence analysis Nonparametric statistics Confidence ellipses Eastern Australia 1. Introduction Silcrete was the most common and widespread lithology used by Australian Aborigines for making flaked stone tools (e.g. Hughes et al., 1973; Sullivan and Simmons, 1979; White and O'Connell, 1982; Hiscock, 1993; Webb et al., 2013; Holdaway and Fanning, 2010; Doelman et al., 2015). This was due to its often high quality flaking properties (Webb and Domanski, 2008) and its abundance. Silcrete oc- curs across almost the entire Australian continent as two distinct geo- graphical associations: in the inland arid regions (e.g. Wopfner, 1978; Thiry and Milnes, 1991; Doelman, 2008), and throughout the more humid parts of eastern Australia (e.g. Young, 1985; Webb and Golding, 1998) where it mostly outcrops close to basalts (Langford-Smith, 1978). In Australia there has been a long-standing perception that silcrete artefacts cannot be accurately traced back to their source, because silcrete types are difficult to distinguish in hand specimen and silcrete is elementally impoverished, making geochemical sourcing problematic. However, silcrete contains significant levels of a variety of trace elements (Webb and Golding, 1998), and geochemical sourcing has been successfully undertaken in southern Africa (Nash et al., 2013a, 2013b). Furthermore, recent advances in analytical technology, particu- larly the widespread availability of Portable X-Ray Fluorescence (PXRF), have made sourcing studies easier. PXRF is in many ways ideal for silcrete sourcing, because it is non-destructive, portable, and can ana- lyse the major and trace element composition of solid samples, includ- ing in situ, so that large numbers of analyses can be rapidly obtained, even in the field (e.g. Goodale et al., 2012; Nazaroff et al., 2010; Sheppard et al., 2010, 2011). PXRF has been successfully used for lithic sourcing of obsidian (e.g. (Torrence et al., 2013), because individual sources of obsidian tend to be chemically homogeneous and therefore distinctive (Craig et al., 2007). However, detection limits, spectral inter- ferences and analytical artefacts mean that perhaps only 7–12 elements can be reliably measured by PXRF (Grave et al., 2012), and there are dif- ficulties with analysis of heterogenous samples. Nevertheless, these limitations can be overcome by employing multiple analyses of individ- ual samples and non-parametric statistics (as discussed below). Archaeological investigations of Aboriginal sites in Australia are re- quired by law to be conducted in consultation with traditional owners, who are frequently interested in lithic sourcing studies but prefer Journal of Archaeological Science: Reports xxx (2016) xxx–xxx ⁎ Corresponding author. E-mail addresses: gcochrane221@gmail.com (G.W.G. Cochrane), john.webb@latrobe.edu.au (J.A. Webb), trudy.doelman@sydney.edu.au (T. Doelman), phabgood@archaeo.com.au, p.habgood@uq.edu.au (P.J. Habgood). JASREP-00712; No of Pages 8 http://dx.doi.org/10.1016/j.jasrep.2016.11.032 2352-409X/© 2016 Elsevier Ltd. All rights reserved. Contents lists available at ScienceDirect Journal of Archaeological Science: Reports journal homepage: www.elsevier.com/locate/jasrep Please cite this article as: Cochrane, G.W.G., et al., Elemental differences: Geochemical identification of aboriginal silcrete sources in the Arcadia Valley, eastern Australia, Journal of Archaeological Science: Reports (2016), http://dx.doi.org/10.1016/j.jasrep.2016.11.032