SCIENTIFIC CHALLENGES
In the Australasian region, palaeoclimatic change has until recently been based on proxy data (in particular pollen) described on a qualitative basis. Although such an approach has some value in providing a general scheme of events, there are inherent problems including the interpretation of proxy data, disentangling different climate signals, temporal sensitivity of proxies to climatic change and the value of qualitative terms.
Quantified reconstructions of summer temperature using fossil Coleoptera remains and their known present-day distribution in the North Hemisphere provide an absolute record of climatic change (Atkinson et al., 1987; Coope et al., 1998), and indicate that absolute temperature changes across northern Europe were considerably more complex than the relative scheme suggests (Witte et al., 1998).
Since this pioneering work, similar quantified estimates have been obtained in the North Atlantic region from Chironomidae (non-biting midges) (Brooks and Birks, 2000), pollen (Nakagawa et al., 2002) and terrestrial plant δ13C (Beerling, 1996). Within the Australasian region, these approaches have had limited application, partially due to the need for the development of contemporary transfer functions. Promising results, however, have been generated for Coleoptera (Porch & Elias, 2000), Chironomidae (Dimitriadis & Cranston, 2001), pollen (Hall and McGlone, 2001) and plant δ13C (Turney et al., 1999).
Determining the timing of Last Termination climate change in terrestrial records has principally been undertaken through radiocarbon dating. The robustness of these chronologies, in terms of precision and accuracy, however, depends upon a number of considerations including the number of radiocarbon ages available for each sequence and the nature of the samples dated (bulk sediment, selected fossils or chemical fractions) (Turney et al., 2000; Lowe et al., 2001).
In many instances within the Australasian region, only a small number of radiocarbon ages have been obtained for Last Termination sequences (largely from bulk sediment samples) and the standard errors on the ages are relatively large (>100 years at 1σ). As a result, it is difficult to assess the accuracy of the ages, isolate aberrant results (resulting for example from in situ taphonomic or biogeochemical processes, field and/or laboratory contamination) and to obtain realistic calibrated age estimates. Ideally, terrestrial plant macrofossils should be utilized as these directly reflect the atmospheric 14C content.
Although single age estimates falling within plateaux may calibrate to span several centuries, a method that offers considerable promise is the use of ‘wiggle-matching' of radiocarbon data sets to the global radiocarbon calibration curve. The methodology is based on the principle that inflections and wiggles in the calibration curve (plateaux and steep transitions) must be reflected in time-depth fluctuations in radiocarbon ages obtained from stratified sequences, and that the latter can be matched to the former to derive calendar ages for the radiocarbon-dated horizons.
