Degree Name

Doctor of Philosophy


School of Geosciences


The study of landforms is a continuous and complex endeavour. The descriptions 'ancient landforms' by Oilier (1991), 'antiquity of landforms' (Twidale, 1998), 'inherited landscapes' by Migoh and Goudie (2001), 'persistent landscapes' of Brunsden (1993) or the 'palimpsest of landform' by Starkel (1987), highlight the diversity and depth of research in the Earth Sciences dedicated to the understanding of ancient landscapes.

The scouring effect of glaciations and periglaciations has largely been believed to have removed any semblances of pre-Quaternary landforms in the Northern Hemisphere. However, as in the Southern Hemisphere, weathering profiles dating back to the Mesozoic have been preserved in places. The recognised juxtaposition or palimpsest nature of landforms, where old and new remnants lie close to each other has resulted from several studies. The methods employed in studying these ancient landscapes should be robust enough to cover the whole timeframe of geological landform history. This study employed the oxygen-isotope technique in studying weathering profiles, supported by palaeomagnetism where appropriate. Both these techniques date weathering profiles that extend well beyond the Quaternary.

The regolith of southeastern and western Australia, in the states of South Australia and Western Australia (Yilgarn Craton) were investigated. This was done based on the pioneering work of Bird and Chivas (1988, 1989 and 1993), where an oxygen-isotope geochronology for the Australian regolith was established. Weathering ages from the regolith covered several geological times, the oldest being pre-late Mesozoic, and the youngest, post-mid Tertiary, which was the most abundantly preserved.

Samples from southern Africa (South Africa) were subjected to a comparative analysis using the Australian geochronology. Isotopic compositions for the various African Erosion Surfaces differ, implying diversity in weathering ages. Some preliminary comparisons with the oxygen-isotope geochronology of Australia showed agreement. The findings are quite encouraging, and warrant further research.

A comparison between palaeomagnetism and oxygen-isotope in dating weathering profiles was assessed. Palaeomagnetism uses the iron-rich material, whereas the oxygenisotope technique largely uses kaolinite. Areas of agreement in ages have been observed, although there were cases of discrepancies, due mainly to resetting of the palaeomagnetic signal.

Palaeomagnetism was used to calibrate the oxygen-isotope signal in iron oxides, mainly hematite within the weathering mantle. The aim was to have a geochronology similar to that established using kaolinites. The findings have not been conclusive, and more detailed sampling is required.

A study of deep weathering profiles sought to explore the nature of weathering as a function of depth - from top to bottom, and in some cases, a lateral aspect (same mine pit depth, but on opposite sides). The lateral aspect has been shown to have little effect on isotopic compositions - that is, samples from either side of a pit yielded similar isotopic compositions. In the Yilgarn Craton, oxygen-isotope variation with depth has been observed in some cases, but not others. Where the weathering profile incorporates distinct weathering phases, the oldest phase occurs deeper in the profile.

The major findings from this study have been: the dating of breakaways in the Yilgarn Craton, some as old as the late Palaeozoic to pre-late Mesozoic (Old Plateau?) and a widespread post-mid Tertiary weathering phase, particularly in its southern portion where rocky outcrops dominate. The weathering ages ofthe breakaways do not fall into a single weathering age, with some equated to the widespread post-mid Tertiary age (New Plateau?). The antiquity of landforms for Australia and southern Africa was highlighted by similarity in geological and oxygen-isotope weathering ages. This requires further investigation. Palaeomagnetism and oxygen-isotope techniques are effective in elucidating landscape evolution, as demonstrated for several Australian sites.

However, it is revealed that both the oxygen-isotope and palaeomagnetic data may be perturbed and re-set by later weathering events. Such complexity, while currently potentially confounding, should prove valuable if the physical and chemical stability of both iron oxides and kaolinite were better understood.