Pleistocene settlement in the Australian arid zone: occupation of an inland riverine landscape in the central Australian ranges

Abstract

Recent excavations at the Kulpi Mara Rockshelter in the Palmer River catchment of central Australia have produced radiocarbon determinations spanning an archaeological sequence of 30,000 years. These results enable re-assessment of models addressing the how, where and when of arid zone colonisation, and human adjustments to environmental change in the later Pleistocene. Whilst the evidence supports early occupation of the central arid zone during wetter conditions, doubts are raised about the continuity of occupation during the height of glacial aridity.

A continental narrative: Human settlement patterns and Australian climate change over the last 35,000 years

“…Mega-lakes were particularly common prior to 48ka, after which most records suggest they entered a sustained drying phase that culminated in contemporary playa conditions with limited input from river systems (Cohen et al, 2015). In parallel to these changes, hunter-gatherers had spread across all parts of the interior by 35ka ( Figure 5), with archaeological sites documented in the Pilbara (Juukan-1 and Djadjiling rockshelter) (Slack et al, 2009;Law et al, 2010), Great Sandy Desert (Puritjarra rockshelter, Parnkupirti) (Smith, 2006(Smith, , 2013Veth et al, 2009), MacDonnell Ranges (Kulpi Mara rockshelter) (Thorley, 1998), Nullarbor (Koonalda Cave) (Wright, 1971), Carnarvon/Geraldton Sand Plains (Mandu Mandu, C99 and Jansz rockshelters) (Morse, 1993;Przywolnik, 2005) and the Swan Coastal Plain (Upper Swan Site) (Pearce and Barbetti, 1981).…”

“…However, the corrected dataset probably provide a more realistic picture of boom and bust populations through this period of high variability in a resource poor environment (see also . The decline at 3ka, for example, is evident as a reduction in artefact discard at Puritjarra rockshelter (base of unit 1b), Kulpi Mara rockshelter (base of layer 1), and a paucity of artefacts at Bush Turkey 3 rockshelter in the Little Sandy Desert (spits 14-19) (Thorley, 1998;Smith, 2006;Veth et al, 2008). This period saw the introduction of numerous technological innovations, most likely resulting from a combination of demographic packing (in-filling of the landscape through population density increases) and long use-life implements such as the Tula adze during the early-to mid-Holocene, restricting residential mobility patterns employed previously in the Pleistocene.…”

Late Quaternary landscapes in Central Australia: sedimentary history and palaeoecology of Puritjarra rock shelter

“…Puritjarra rock shelter is one of several sites with long (>20 ka) archaeological sequences in the interior of the Australian arid zone (Veth, 1995;O'Connor et al, 1998;Thorley, 1998;Smith, 2005b). Archaeological excavations between 1986 and 1990 showed that the sedimentary history of the site spans the last 100 ka, with most evidence of human occupation falling within the last 35 ka.…”

“…Elsewhere in central Australia, a range of palaeoenvironmental data has come from linear and source-bordering dunes (Nanson et al, 1992(Nanson et al, , 1995Twidale et al, 2001;Hollands et al, 2006;Fitzsimmons et al, 2007a), playas, evaporites and groundwater systems (Chen et al, 1991(Chen et al, , 1993English et al, 2001), and floodplain and palaeoflood deposits (Pickup et al, 1988;Patton et al, 1993;Tooth, 1997;Bourke, 1998) but these have been used principally as proxies for palaeoclimate, rather than integrated reconstructions of late Quaternary landscapes. In contrast, the growing body of archaeological research in Australia's arid zone has sharpened the need for fine-grained reconstruction of regional palaeoenvironments, as a prerequisite for evaluating patterns of humanenvironment interaction, resource use and settlement histories (Smith, 1989(Smith, , 2005bVeth, 1993Veth, , 1995Hiscock, 1994;Thorley, 1998;Ross, 2003;Hiscock and Wallis, 2005).…”

“…Wallis, 2001), this type of evidence has not been systematically exploited in Australia since the 1970s. However, in Australian deserts, the scarcity of other palaeoecological records underscores the potential of Late Pleistocene deposits at archaeological sites, such as Kulpi Mara (Thorley, 1998) and Serpent's Glen (O'Connor et al, 1998), for this type of research.…”