Australia: The Land Where Time Began

A biography of the Australian continent 

The Last Glacial Period – Climatic and Environmental Changes 30-20 Cal. ka BP

Reconstructions that are continuous and well-dated through at least part of this period are provided by 3 pollen records. From Lake Euramoo in the Wet Tropics World Heritage rainforest of northeast Queensland a new, extended pollen and charcoal record has been developed (Haberle, 2005). Incorporated in the 8.4 m sediment core that was taken from the centre of Lake Euramoo is a complete vegetation change record and fire history that spans the period from 23 cal. ka BP (the time at which the crater formed) to the present. The vegetation was dominated by sclerophyll with a peak of herbs during the last glacial. To the immediate south of Lynch’s Crater a virtually identical oscillation with increased representation of herbs is recognised (Kershaw, 1986; Turney et al., 2006), which represents a contraction of rainforest taxa, and is inferred to be a correlative of the conditions that were cold and dry during the LGM. It was indicated by a reconstruction from Tower Hill, Victoria, that was published previously, to be a record of taxa dominated by steppe-grassland from at least 22 cal. ka BP (D’Costa et al., 1989). Taken together, the LGM as recorded by changes of vegetation at these 3 sites appears to have spanned the period 23 and 18 cal. ka BP.

It has been revealed by TL dating of aeolian activity near Birdsville, Queensland, on the eastern edge of the Simpson Desert, and near Finke in the Finke River Valley, the Northern Territory, on the western edge of the Simpson Desert (Nanson et al., 1995), aspects of the Late Pleistocene history of the great anticlockwise whorl of the dunes that cover much of central Australia. The linear dunes that are oriented northwesterly at Birdsville have palaeosols that are well defined, and are almost certainly older than the basal ages of about 80 ka that have been obtained so far, though they have been reworked extensively since the LGM, during the Holocene in particular (Nanson et al., 1992a, 1992b). The regional dunefield at Finke consist of linear dunes that are dark red, and that are aligned essentially due north, before and after the LGM and Termination, 30-12 ka. Dunes of dark red fine sand older than 70 ka have been described at Camel Flat in the northwest Simpson Desert and are, as at Finke, oriented almost due north (Hollands et al., 2006).

Coupled to this anticyclonic airflow in the interior of the continent, it has been reported (Miller et al., 1997) the temperature-dependent component of the amino acid racemisation reaction in fragments of emu eggshells dated by radiocarbon. It was found that the racemisation rate changes were consistent with an average air temperature of at least 9^oC cooler during the last glacial period, from at least 40 ka to about 16 ka, that was possibly related to a reduced atmospheric moisture content at the time.

Contrasting with contemporary Australia’s interior that is largely arid, it is shown by the alluvial record that there were episodes when the continent was drained by large, powerful rivers that filled Lake Eyre (Nanson et al., 1992a). The record of lake level from Lake Eyre shows higher, and probably levels that were oscillating, around 40 to 30 ka, though during the LGM a dry playa (Magee et al., 2004). On the Murrumbidgee Riverine Plain, southeastern Australia, fluvial deposits that are the most thoroughly dated in Australia, show several periods during which palaeochannel activity was greatly enhanced.  The Gum Creek phase spanned from 35 to 25 ka. Contrasting with this the latter part of the glacial period, 25-20 ka, appeared to be one of reduced fluvial activity, which is consistent with conditions that were more arid at this time (Nanson et al., 2003).

New chronologies of advance and retreat of glaciers through this period are being provided by exposure dating of moraine sequences at Mt Kosciuszko in the Snowy Mountains on the mainland of Australia (Barrows et al., 2001) and in Tasmania in alpine valleys (Fink et al,, 2000; Barrows et al., 2002; Kiernan et al., 2004, and unpublished data). At Mt Kosciuszko the maximal glacial advance was centred on 19.6 ± 1.6 [cal. ka] (Blue Lake Advance) while periglacial activity, that was apparently synchronous, was dated to 21.9 ± 0.5 cal. ka, which reflects cooling of 9-11^oC (Galloway, 1965); Barrows et al., 2004). According to Turney et al. the glacial record that is emerging from work in Tasmania at Lake Rolleston, Cradle Mountain, Mt Murchison suggest maximum glacial advances that are associated with the last glacial period at 23 ± 2 and 29 ± 2 ka, though the strength of the advances during the LGM appear to be limited compared to those in the records from the Northern Hemisphere (Fink et al., unpublished data). Though it is at present uncertain what the climatic drivers of these events are there were clearly changes that were significant and sustained in temperature and/or precipitation (in the form of snowfall in winter) to drive major changes in the net balance of glaciers through time.

The most comprehensive study to date within the ocean realm has been for the time-slice of the LGM, which was dated by the use of radiocarbon and oxygen isotope stratigraphy (Barrows & Juggins, 2005). Here, sea surface temperature (SST) reconstructions that are based on planktonic assemblages of foraminifera demonstrate that the coldest period was centred on 20.5 ± 1.4 cal. ka BP. Turney et al say this predated the maximum in oxygen isotope records at 18.2 ± 1.5 cal. ka BP. In the tropics cooling was between 0 and 2^oC, and along the western coast of Australia there was a maximum depression of temperature of 4^oC. The water of the temperate southwest Pacific cooled by 3-5^oC, the largest sea surface temperature anomalies occurring to the south of latitude 40^oS, and temperatures were as low as 8^oC lower than at present. During the LGM the Indo-Pacific Warm Pool was significantly smaller, and centred to the northeast of its present location, while the Leeuwin Current and East Australian Current were operating, though they were transporting cooler waters than at the present. It appears the South Pacific Convergence Zone was absent at the time. It also appears the northward expansion of the Southern Ocean contributed to the steeper sea surface temperature gradient relative to the present day.

Evidence of climate change in the source areas of the northwestern and southeastern arid zone is provided by studies of dust in marine sediments around Australia (Hesse and McTainsh, 2003). The glacial dust flux within core E26.1 was up to 7 times higher during minimum Sea Surface Temperatures (inferred to be the LGM). A northward shift of about 3^o was associated with this dust deposition increase. Turney et al., say the implied northward shift in the summer westerly circulation and greater wind erosion is in broad agreement with the increase of aridity that is inferred from the activity of dunes, reduction of fluvial activity and the steeper Sea Surface Temperature gradient observed in the Southern Ocean. Turney et al. suggest that the discrepancy between the timing of the LGM as was recorded in E26.1, which was centred on 25 cal. ka BP, and elsewhere is probably the limited number of radiocarbon dates for the marine sequence and the uncertainty of reservoir ages through this period. It has been revealed by particle size studies of the Tasman dust (Hesse & McTainsh, 1999), that wind speeds during the LGM were comparable to those of the present and that in source areas mechanisms of supply and deposition were the dominant control of the dust flux. The glacial dust flux, in contrast, was about 20 ka was significantly lower in the Indian Ocean core of SO14-08-05, though of coarser particle size than was the case in the Holocene. This followed at about 30 ka an interval during which dust deposition was higher. It is still unclear what the cause of this significant degree of variability could have been.

The accumulation rate of ice at the LGM that is recorded in the Antarctic record from Law Dome (van Ommen et al., 2004) is estimated to have been less than 10 % of the modern value. Sustained cooling throughout this part of the record is inferred to have been represented by depleted ¹⁸O values. A phase of maximum cooling that was comparable in timing to the LGM has not been found on the mainland of Australia.

Sources & Further reading

1. Turney, C. S. M., S. Haberle, D. Fink, A. P. Kershaw, M. Barbetti, T. T. Barrows, M. Black, T. J. Cohen, T. Corrège, P. P. Hesse, Q. Hua, R. Johnston, V. Morgan, P. Moss, G. Nanson, T. van Ommen, S. Rule, N. J. Williams, J. X. Zhao, D. D'Costa, Y. X. Feng, M. Gagan, S. Mooney and Q. Xia (2006). "Integration of ice-core, marine and terrestrial records for the Australian Last Glacial Maximum and Termination: a contribution from the OZ INTIMATE group." Journal of Quaternary Science 21(7): 751-761.