Human refugia in Australia during the last glacial maximum and terminal pleistocene: A geospatial analysis of the 25-12ka Australian archaeological record

A number of models, developed primarily in the 1980s, propose that Aboriginal Australian populations contracted to refugia - well-watered ranges and major riverine systems - in response to climatic instability, most notably around the Last Glacial Maximum (LGM) (~23-18ka). We evaluate these models using a comprehensive continent-wide dataset of archaeological radiocarbon ages and geospatial techniques. Calibrated median radiocarbon ages are allocated to over-lapping time slices, and then K-means cluster analysis and cluster centroid and point dispersal pattern analysis are used to define Minimum Bounding Rectangles (MBR) representing human demographic patterns. Exploring data between 25 and 12ka, we find a refugia-type hunter-gatherer response during the LGM (~23-18ka) and again during the Antarctic Cold Reversal (ACR) (~14.5-12.5ka), with expansion in the intervening period. Several refugia persist between 25 and 12ka, including (by Interim Biogeographic Regionalisation for Australia areas) Gulf Plains/Einasleigh Uplands, Brigalow Belt South, Murray Darling Depression, and Tasmanian Central Highlands. Others appear sporadically through the same period. These include South Eastern Highlands, NSW South Western Slopes, Sydney Basin, Warren, Murchison, Gascoyne, Central and Northern Kimberley, Ord Victoria Plain, Arnhem Plateau, MacDonnell Ranges and Central Ranges. The Pilbara may also have been a refugium during the LGM, but geospatial results are problematic for this region. Areas devoid of human activity ('barriers') include the main desert regions, especially in the south and west of the continent, although some of these may be the result of an absence of archaeological fieldwork. Point dispersal pattern analysis indicates a reduction in occupied territory of nearly 80% during the LGM. A reduction of close to 50% was also evident during the ACR. A large number of the refugia were in close proximity to glaciated areas during the LGM, and probably benefitted from increased summer snowmelt along the major river systems. The remaining refugia are likely the result of a range of local environmental and resource factors. We identify areas for future research, including a focus on regional studies to determine possible cryptic or idiosyncratic refugia emerging in phlyogeographic studies.