[Truncated abstract] Establishment of vegetation and ecosystem functioning is central to the mitigation of environmental impacts associated with mining operations. This study investigated the ecophysiological functioning of mature plants in natural vegetation and applied this knowledge to diagnose problems affecting plant health and causes of poor plant cover at a mine-rehabilitation site. Ecophysiological parameters, including plant water relations and mineral nutrition, were studied in conjunction with soil physical, hydraulic and chemical properties. The natural ecosystem at the study location in the Great Sandy Desert is characterised by sand dunes and interdunes with distinct plant communities on each. One of the most notable features of the vegetation is the presence of large Corymbia chippendalei trees high on the dunes and relatively small scattered shrubs in the interdunes. Triodia grasses (spinifex), dominate the vegetation in both habitats but different species occur in each; T. schinzii is restricted entirely to the dunes, and T. basedowii occurs only in the interdunes. It was hypothesised that the deep sandy dunes afford greater water availability but lower nutrient supply to plants in this habitat compared with those occurring in the lower landscape position of the interdunes. Water-relations parameters (leaf water potentials, stomatal conductance, d13C) revealed that dune plants, particularly woody species, displayed higher water status and water use than closely related and often congeneric plants in the interdunes. Nutrient concentrations in soils were significantly higher in the interdunes, but concentrations in foliage were similar for related species between habitats. It is concluded that the dunes provide a greater store of accessible water than the soil profile in the interdunes. ... Following an experimental wetting pulse equivalent to a summer cyclone event, A. ancistrocarpa plants displayed significant increases in stomatal conductance, leaf water potential and sap velocity in lateral roots within three days of irrigation at the natural site and two days at the rehabilitation site. Secondary sinker roots originating from distal sections of lateral roots were evidently supplying water to maintain hydraulic function in laterals, thus enabling a fast pulse response. This was accentuated at the rehabilitation site where roots were confined closer to the surface. These results indicate that plants at the rehabilitation site are more dependent on small pulses of water and have less access to deep reserves than plants at the natural site. It is concluded that high runoff losses and insufficient soil depth are major factors contributing to plant water stress, and combined with the direct impacts of erosion, are largely responsible for plant death and ultimately poor plant cover. These issues can be alleviated if cover soil depth is increased to more than 0.5 m and slope angles are reduced to <12o. This study demonstrates the value of an ecophysiological approach for diagnosing problems affecting plant establishment at mine-rehabilitation sites. Furthermore, it has provided recommendations that will improve the rehabilitation strategy and lead to the development of a well vegetated, resilient ecosystem on a stable and non-polluting land form.

[Truncated abstract] This ethnoecological study examines land uses by modern Martu Aboriginal people on their country. They occupy very remote settlements—Parnngurr, Punmu and Kunawarritji—in the Great and Little Sandy Deserts. In 1990, their country included Crown Lands and Rudall River National Park. The study investigated the proposition that the knowledge and practices of Martu were of direct relevance to ecosystem processes and national park management. This research commenced in the wider Australian research context of the late 1980s – early 90s when prevailing questions were about the role of customary harvest within contemporary Aboriginal society (Altman 1987; Devitt 1988) and the sustainability of species-specific harvests by Australian indigenous people (Bomford & Caughley 1996). Separately, there was a national line of enquiry into Aboriginal roles in natural resource and protected area management (Williams & Hunn 1986; Birckhead et al. 1992). The field work underpinning this study was done in 1986–1988 and quantitative data collected in 1990 whilst the researcher lived on Martu settlements. Ethnographic information was gathered from informal discussions, semi-structured interviews and participant observation on trips undertaken by Martu. A variety of parameters was recorded for each trip in 1990. On trips accompanied by the researcher, details on the plant and animal species collected were quantified. Martu knowledge and observations of Martu behaviour are interpreted in terms of the variety of land uses conducted and transport strategies including vehicle use; the significance of different species collected; socio-economic features of bush food collection; spatio-temporal patterns of foraging; and, the 'management' of species and lands by Martu. The research found that in 1990, hunting and gathering were major activities within the suite of land uses practiced by Martu. At least 40% of trips from the settlements were principally to hunt. More than 43 animal species and 37 plant food species were reported to be collected during the study; additionally, species were gathered for firewood, medicines and timber artefacts. Customary harvesting persisted because of the need for sustenance, particularly when there were low store supplies, as well as other reasons. The weight of bush meats hunted at least equalled and, occasionally, was three times greater than the weights of store meats available to Parnngurr residents. ... Paradoxically, hunting was a subject of significant difference despite it being the principal activity driving Martu expertise and practice. There is potential for comanagement in the National Park but it remains contingent on many factors between both Martu and DEC as well as external to them. The dissertation suggests practical strategies to enhance co-management.