Late Quaternary erosion events in lowland and mid-altitude Tasmania in relation to climate change and first human arrival

Abstract The establishment of a chronology of landscape-forming events in lowland and mid-altitude Tasmania, essential for assessing the relative importance of climatic and human influences on erosion, and for assessing present erosion risk, has been limited by the small number of ages obtained and limitations of dating methods. In this paper we critically assess previous Tasmanian studies, list published radiocarbon ages considered to be dependable, present new radiocarbon and thermoluminescence (TL) ages for 25 sites around Tasmania, and consider the evidence for the hypotheses that erosion processes at low and mid altitudes have been: (1) purely climatically controlled; and (2) influenced both by climatic and anthropogenic (increased fire frequency) effects. A total of 94 dependable finite ages (calibrated for radiocarbon and ‘as measured’ for TL and optically stimulated luminescence (OSL) determinations) are listed for deposits comprising dunes, colluvium, alluvium and loess-like aeolian deposits. Two fall in the >100 ka period, 15 fall in the period 65–35 ka, and 77 fall in the period 35–0.3 ka. There was a sustained increase in erosion recorded in the period 35–15 ka, as reflected by a greater number of dated aeolian deposits during this period. We considered three possible biases that may have affected the age distribution obtained: the limitations of radiocarbon dating, sampling bias, and preservation bias. Sampling bias may have favoured more recent dune strata, but radiocarbon dating and preservation biases are unlikely to have significantly distorted the age distribution obtained. Long but intermittent aeolian deposition is recorded at two sites (Southwood B; c. 59–28 ka and Dunlin Dune; c. 29–14 ka) but there is no evidence of regional loess deposits such as found in New Zealand. The timing of increased erosion in Tasmania between 35 and 30 ka approximately coincides with the intermittent ten-fold increase of dust accumulation between 33 and 30 ka in the Antarctic Dome C ice core. The absence of widespread erosion before 35 ka, the abrupt increase of erosion around this time, the frequent association of erosion products with charcoal, the arrival of people in Tasmania at c. 40 cal ka, and the known use of fires by Aborigines to maintain areas of non-climax vegetation suggest that ecosystem disturbance by anthropogenic fires, in a drier climate than that presently prevailing, may have contributed to erosion in lowland and mid-altitude Tasmania after 35 ka. Thus the Tasmanian erosion record provides circumstantial support for the proposition that human dispersal in southeast Australia was accompanied by significant ecological change.