Steep environmental gradients, strong correlations between temperature and rainfall, and the potential impact of variation in CO2 concentrations complicate the interpretation of past climate change in eastern Australia. Physical or biological proxies with independent responses to temperature, rainfall, and atmospheric content would be ideal candidates for deriving quantitative estimates of past climate. These seldom exist or are widely applicable. The ecological diversity of beetles and the fact they can be specifically identified negates several of these inherent limitations. The distribution of individual beetle taxa is determined by a multitude of parameters, however, it has been demonstrated that for most predators, scavengers and generalist herbivores, climatic parameters are paramount.
I will outline the development of a taxon-based method for the reconstruction of Quaternary climates in Australia using fossil beetles. For extant assemblages, the method enables reliable and accurate estimation of summer and winter temperatures and several rainfall parameters. The accuracy of the estimates, when compared to modern climatic data, depends on the number of taxa in the assemblage having bioclimatic profiles, and the presence of stenotopic ëindicatorí taxa. Estimates of mean summer and winter temperatures to within ±1.5°C, and mean annual rainfall to within ±150mm are routinely achieved. A late Pleistocene fossil assemblage from Spring Creek in western Victoria is described to illustrate the application of the method the Quaternary deposits. The Spring Creek deposit, dated to >40 kyr. B.P., contains a diverse aquatic and riparian assemblage similar to those occurring on the western volcanic plains today, but containing a number of taxa not presently found in the region. Reconstruction of the climate at the time of deposition of the Spring Creek fauna indicate cooler winter temperatures, a summer climate similar to today, and a similar, perhaps slightly greater annual rainfall. Aquatic caddis assemblages reveal similar temperature regimes. Significantly, pollen assemblages from Spring Creek are essentially ëglacialí in character, dominated by grasses and daisies, with small numbers of Casuarinaceae, very few Myrtaceae, and small but consistent amounts of Pimelia, Plantago, Myriophyllum, and Ranunculaceae.
Palaeoclimatic reconstruction using fossil beetles is limited by few of the factors that apply to the quantitative reconstruction of past climates using pollen assemblages. In Australia it is, however, less generally applicable than pollen analysis because of the difficulties in obtaining fossil bearing material in sufficient quantities, especially from extended sequences. Nonetheless, beetle-based climatic reconstructions from small slices of the Quaternary can be compared with associated pollen data to assess and refine pollen-climate relationships.