Thesis abstract: Palaeo-Environmental Change and the Persistence of Human Occupation in South-Western Australian Forests

Centre for Archaeology, University of Western Australia; University of Western Australia Library; Australian Institute of Aboriginal and Torres Strait Islander Studies Library; Department of Archaeology and Natural History, Research School of Pacific and Asian Studies, Australian National University; Western Australian Museum; Department of Indigenous Affairs (WA) Library; Calgardup Cave Interpretation Centre, Department of Conservation and Land Management (WA)

This thesis investigates hunter-gatherer responses to environmental changes in south-western Australian forests. Specifically, it examines how hunter-gatherers reacted to terminal Pleistocene and early Holocene expansions of Karri (Eucalyptus diversicolor) tall open-forest, a forest type identified as difficult to occupy. The thesis presents archaeological evidence for hunter-gatherer occupation of sites in different forest types within south-western Australia which show that hunter-gatherers persisted in occupying the whole forested region throughout the vegetational changes. They did this by following small geographical shifts in favourable habitats and controlling the extent of unfavourable habitats by firing.

My approach is to compare environmental histories from archaeological sites with records to site occupation extending from the last few millennia to 47,000 BP. The study area, the Leeuwin-Naturaliste Region, south-western Australia, is chosen for its deep limestone cave deposits providing long records of human occupation and environmental change, and for its intricate mosaic of forest communities. In this region, Devil's Lair, Tunnel Cave, Witchcliffe Rock Shelter and Rainbow Cave have stratified sandy floor deposits containing hearths, stone artefacts, and biotic remains. The middle two sites of this series were excavated in my own systematic survey for cave sites. Devil's Lair and Tunnel Cave are located in present-day Karri forest, Witchcliffe Rock Shelter and Rainbow Cave in coastal woodland and scrub.

Radiocarbon assays and stratigraphy indicate that at Devil's Lair, episodes of human occupation extended from 47,000 BP until the entrance collapsed, sometime after 12,000 BP. At Tunnel Cave, there are six hearth complexes built between 20,000 and 12,000 BP, traces of occupation up to 8000 BP, and a hearth at 1,400 BP. Witchcliffe Rock Shelter and Rainbow Cave were each occupied by hearth-building people, 800-400 BP.

Analysis of trends towards conserving raw material in stone artefact manufacture and use suggest no changes in occupation intensity at any site, except at Tunnel Cave during the height of the last glacial, when a thick hearth layer was built up. Cold and wind at this time perhaps encouraged more frequent cave occupation, but no climatic change could have required people to abandon Tunnel Cave as a campsite from 8000 to 1400 BP. A more likely factor in cave occupation/abandonment in the Holocene was change in the vegetation surrounding cave sites.

A change in vegetation structure or habitat is indicated by the proportions of mammal species in bone fragments (excluding hunter-gatherer prey animals deposited out of proportion to their natural representation). Identified charcoal fragments indicate changes in the floristic composition of canopy dominants. these analyses show that rainfall increased and habitat became more closed from 13,000 BP, as the Pleistocene Jarrah forest or woodland gave way to Karri forest. At Tunnel Cave, the latter formation had encroached totally by 8000 BP, the same time that people abandoned the site.

These results show primarily that hunter-gatherer site occupation altered in response to vegetational change. However, previous research cited in the thesis shows that people did not abandon the Leeuwin-Naturaliste Region when Karri forest encroached over some parts and people did return to some sites in Karri forest. As in historical times, forest occupation probably involved flexible, short-term site occupations, and therefore the slight vegetational shifts, had little impact on regional occupation patterns. The wider implications are that hunter-gatherers prioritised their use of different vegetation communities, and in all regions where vegetation communities form a mosaic, hunter-gatherers can maintain the same occupation pattern by following geographic shifts of those communities.