ARTICLE Post-Last Glacial Maximum settlement of the West Angelas region in the inland Hamersley Plateau, Western Australia Michael Slack a,b , Kate Connell b,c , Annabelle Davis d , Luke Andrew Gliganic b,e , W. Boone Law b,f and Michael Meyer e a ARC Centre of Excellence for Australia Biodiversity and Heritage, College of Arts, Society and Education, James Cook University, Cairns, Australia; b Scarp Archaeology, Terrey Hills, NSW, Australia; c School of Social Science, The University of Queensland, Brisbane, Australia; d Communities and Communications, Rio Tinto, Perth, Australia; e Institute for Geology, University of Innsbruck, Innsbruck, Austria; f School of Biological Sciences, University of Adelaide, Adelaide, Australia ABSTRACT An excavation and survey program at West Angelas, in the Pilbara region of Western Australia, shows that the poorly watered interior area of the Hamersley Plateau was first occupied soon after the conclusion of the Last Glacial Maximum, and that significant use of this area probably only occurred during the mid to late Holocene. Although current archaeo- logical research shows that Aboriginal groups have occupied areas of the Hamersley Plateau for more than 40,000 years, the permanent and prolonged use of the more marginal or eco- logically suboptimal foraging environments of the interior plateau is a comparatively recent development in the region’s long archaeological record. ARTICLE HISTORY Received 3 November 2016 Accepted 14 October 2017 KEYWORDS Rockshelter excavation; Pilbara; post glacial subsistence and settlement Introduction Research on the late Pleistocene settlement of the Hamersley Plateau has been mainly in two areas: (i) identifying early occupation, now established before 40,000 years ago (Law et al. 2010;O’Connor et al. 1998) and (ii) establishing whether or not the region provided a refuge allowing settlement to ensue dur- ing the Last Glacial Maximum (LGM) (Brown 1987; Hiscock 2008; Slack et al. 2009; Smith 2013; Veth 1993, 1995, 2005; Williams et al. 2013). In this paper the LGM is defined as the period between 33,000 and 20,400 years ago, representing the driest period of the last 100,000 years, with no seasonal monsoonal activ- ity or summer rainfall recorded (Fitzsimmons et al. 2012; Nott 2011; Reimer et al. 2013; Williams et al. 2009; Wyrwoll and Miller 2001). Previous discussions have concentrated on the handful of archaeological sites that provide a record of this period such as Newman Rockshelter (Troilett 1982), Newman Orebody XXIX (Maynard 1980), Milly’s Cave (Marwick 2002), Juukan (Slack et al. 2009) and Djadjiling (Law et al. 2010). Most of these sites occur along substantial river systems and along the edges of the plateau. However, outside the Hamersley Plateau, other Pilbara sites have been argued to straddle the LGM (Morse et al. 2014; Williams et al. 2015). In addition to a focus on the earliest sites of the peripheral Hamersley Plateau, there have been many large-scale site based landscape studies in the region, particularly in response to mining development. Most of the sites excavated as part of this work do not include a pre-30,000 year component. In fact, as noted by Morse (2009:2), over 80% of 126 dates available from the inland Pilbara are less than 4,000 years old, with many of the results largely confined to what is known as the ‘grey literature’, chiefly con- sultancy reports. In exceptions, however, Ryan and Morse (2009) examined the characteristics of stone artefact scatters in order to propose a model of Holocene site use and Bird and Rhoads (2015) looked at rockshelters and whether they can be clas- sified as ephemeral sites, as they usually are. Here, we build on Ryan and Morse’s consultancy- based archaeological modelling to propose that the post-LGM record of the Pilbara, and particularly sites with occupation commencing from the terminal Pleistocene/late Holocene warrant particular atten- tion for what they mean in terms of the past Aboriginal expansion and settlement at a regional to local scale. This is illustrated through an example from West Angelas (Figure 1(a, b)) that contains an archaeological record commencing after the peak of the LGM, but with evidence for intensive occupation only during the mid to late Holocene. In presenting our evidence, we propose a model in which, despite early occupation of the Hamersley Plateau, marginal rangeland environments were possibly avoided and largely unoccupied during the peak LGM period. CONTACT Michael Slack michael.slack@jcu.edu.au ARC Centre of Excellence for Australia Biodiversity and Heritage, College of Arts, Society and Education, James Cook University, PO Box 6811, Cairns, Qld 4870, Australia ß 2017 Australian Archaeological Association AUSTRALIAN ARCHAEOLOGY, 2017 VOL. 83, NO. 3, 127–142 https://doi.org/10.1080/03122417.2017.1404548