Thesis (PhD)--Macquarie University, Division of Environmental & Life Sciences, Graduate School of the Environment, 2002. / Includes bibliographical references: p. 228-232. / Geomorphology, archaeology and geoarchaeology: introduction and background -- Surface stone artefact scatters: why can we see them? -- Geomorphic controls on spatial patterning of the surface stone artefact record -- A temporal framework for interpreting surface artefact scatters in Western NSW -- Synthesis: stone artefact scatters in a dynamic landscape. / Surface scatters of stone artefacts are the most ubiquitous feature of the Australian Aboriginal archaeological record, yet the most underutilized by archaeologists in developing models of Aboriginal prehistory. Among the many reasons for this are the lack of understanding of geomorphic processes that have exposed them, and the lack of a suitable chronological framework for investigating Aboriginal 'use of place'. This thesis addresses both of these issues. -- In arid western NSW, erosion and deposition accelerated as a result of the introduction of sheep grazing in the mid 1800s has resulted in exposure of artefact scatters in some areas, burial in others, and complete removal in those parts of the landscape subject to concentrated flood flows. The result is a patchwork of artefact scatters exhibiting various degrees of preservation, exposure and visibility. My research at Stud Creek, in Sturt National Park in far western NSW, develops artefact and landscape survey protocols to accommodate this dynamic geomorphic setting. A sampling strategy stratified on the basis of landscape morphodynamics is presented that allows archaeologists to target areas of maximum artefact exposure and minimum post-discard disturbance. Differential artefact visibility at the time of the survey is accommodated by incorporating measures of surface cover which quantify the effects of various ephemeral environmental processes, such as deposition of sediments, vegetation growth, and bioturbation, on artefact count. -- While surface stone artefact scatters lack the stratigraphy usually considered necessary for establishing the timing of Aboriginal occupation, a combination of radiocarbon determinations on associated heat-retainer ovens, and stratigraphic analysis and dating of the valley fills which underlie the scatters, allows a two-stage chronology for huntergatherer activity to be developed. In the Stud Creek study area, dating of the valley fill by OSL established a maximum age of 2,040±100 y for surface artefact scatters. The heatretainer ovens ranged in age from 1630±30 y BP to 220±55 y BP. Bayesian statistical analysis of the sample of 28 radiocarbon determinations supported the notion, already established from analysis of the artefacts, that the Stud Creek valley was occupied intermittently for short durations over a relatively long period of time, rather than intensively occupied at any one time. Furthermore, a gap in oven building between about 800 and 1100 years ago was evident. Environmental explanations for this gap are explored, but the paiaeoenvironmental record for this part of the Australian arid zone is too sparse and too coarse to provide explanations of human behaviour on time scales of just a few hundred years. -- Having established a model for Stud Creek of episodic landscape change throughout the late Pleistocene and Holocene, right up to European contact, its veracity was evaluated in a pilot study at another location within the region. The length of the archaeological record preserved in three geomorphically distinct locations at Fowlers Gap, 250 km south of Stud Creek, is a function of geomorphic dynamics, with a record of a few hundred years from sites located on channel margins and low terraces, and the longest record thus far of around 5,000 years from high terrace surfaces more remote from active channel incision. But even here, the record is not continuous, and like Stud Creek, the gaps are interpreted to indicate that Aboriginal people moved into and out of these places intermittently throughout the mid to late Holocene. -- I conclude that episodic nonequilibrium characterizes the geomorphic history of these arid landscapes, with impacts on the preservation of the archaeological record. Dating of both archaeological and landform features shows that the landscape, and the archaeological record it preserves, are both spatially and temporally disjointed. Models of Aboriginal hunter-gatherer behaviour and settlement patterns must take account of these discontinuities in an archaeological record that is controlled by geomorphic activity. -- I propose a new geoarchaeological framework for landscape-based studies of surface artefact scatters that incorporates geomorphic analysis and dating of landscapes, as well as tool typology, into the interpretation of spatial and temporal patterns of Aboriginal huntergatherer 'use of place'. / Mode of access: World Wide Web. / vii, 232 p. ill., maps