THERE HAS BEEN a great deal of environmental change in the Australian region over the 50,000 years that people have lived here. There is going to be a lot more of it in the future, too, whether or not the specifics of current climate-change models are borne out.
About 120,000 years ago, temperatures and sea levels were like today's but then began falling. By the time people first settled the continent 50,000 years ago, sea levels were perhaps thirty metres lower than now, the climate was mild and wet and many now-dry inland lakes were full. The lower parts of the Murray-Darling system provided a particularly rich environment for people. This has been revealed in stunning detail by archaeological research at Lake Mungo in south-western NSW. These days, Mungo is a flat, parched expanse of low scrub edged by the thirty-metre high, thirty-kilometre long and mightily impressive Walls of China sand dune. When the lake was full, there was an abundance of clear water, freshwater fish including Murray Cod, mussels and crustaceans. Emu and a wide variety of marsupials including the "Tasmanian" tiger also flocked to the water. Waterfowl probably did, too, although their bones have not survived. The remains of all the other creatures, as well as more enduring parts of the tools and weapons used to hunt and process them, have been found in ancient shell middens and campfires in the great sand dune, indicating the extensive use people made of the lake and its surrounds tens of millennia ago.[i]
Life at Lake Mungo and elsewhere around Australia at the time was not to last, as environmental change forced dramatic changes. Temperatures, rainfall and sea levels plummeted about 35,000 years ago and continued to fall until the Last Glacial Maximum (LGM) 25,000 to 12,000 years ago. Average sea-surface temperatures during this period were up to four degrees lower than now in the tropics and up to nine degrees lower further south. Sea levels dropped 130 metres below today's levels as the world's water was taken up in the northern hemisphere's vast icesheets.[ii]
Despite the name, there was actually very little ice in Australia during the LGM. In Europe and North America, icesheets kilometres thick relentlessly ground down huge areas of the landscape during this period. Films such as The Day after Tomorrow, much less the Disney cartoon Ice Age, do not even begin to show what it would have been like under full glacial conditions in the northern hemisphere.
Australia's climate, on the other hand, became colder but, more importantly, it also became much more arid. The cold pushed the tree line downhill in highland areas, but the main environmental shift was a major expansion of the continent's arid/semi-arid core. Today, such environments occupy nearly three-quarters of Australia's land area but then, nearly the whole continent was desert or semi-desert.
The spreading of the arid zone was accompanied by dune building on a massive scale. This process left a legacy still visible in the form of large but stabilised and well-vegetated dunes in what are now much wetter parts of the country, including Tasmania. Reflecting the patterns of prevailing winds, the dunes are arranged in a giant swirl across the continent. From space, the swirl looks like an enormous thumb print on the land. At the time of its creation, though, things would often have been even less pleasant than usual during that period, as dust storms increased markedly in frequency and severity.
While all this was happening, the continued lowering of sea levels created land bridges between Australia, New Guinea and Tasmania, forming a super-continent scientists call Sahul. The coastline expanded elsewhere as well, dramatically increasing the size of the continent. Large, brackish lakes formed in what are now the Gulf of Carpentaria and Bass Strait. Climate change and sea-level variation had relatively little impact in New Guinea (then northern Sahul) and adjacent archipelagos, though alpine areas expanded and upland vegetation communities ringing the highlands moved downhill. Sparsely treed grassland like that now seen around Port Moresby probably expanded in southern New Guinea.
The beginning of the end of the last glacial period was marked by a rise in temperature and sea levels about 15,000 years ago, though it remained very cold and dry for about another 3,000 years. Dune building ended and tree lines began migrating back uphill. Sea-level rise broke up Sahul, with Tasmania finally separating about 14,000 years ago and New Guinea about 7,000 years ago. Australia's land area shrank considerably. John Mulvaney and Johan Kamminga[iii] paint a graphic picture of the situation at the time:
Vast territories, innumerable camping places, stone quarries, burial grounds, and sacred features, were submerged; the creation of continental islands isolated populations. In most situations, the inundation of land was barely perceptible within a human lifetime, and individuals were not displaced from their ... territory. But where coastal land was flat ... [sea-level rise] would have been noticeable, and even a cause for concern. On the flat Great Australian Bight and Arafura Shelf between 13,000 to 11,000 years ago, the ... [sea rose at] one metre a week ... Regions larger than modern tribal territories would have disappeared.
These trying times did not end with the passing of the Ice Age, either. At present, the nature of postglacial environmental shifts remains hard to pin down in the southern hemisphere. In the northern hemisphere, there was a well-defined series of arid periods (eg the Younger Dryas 12,800-11,500 years ago), but there is no clear indication that any of them occurred in the southern hemisphere. Much the same can be said of the "Medieval Warm Period" between the ninth and thirteenth centuries AD, when Greenland was vegetated and thus, in fact, green. So, too, with the "Little Ice Age" between the fifteenth and late nineteenth centuries AD, when the Thames regularly froze over.[iv] We do know that sea levels around Australia stabilised at present levels about 6,000 years ago, after a period when they actually rose up to a metre or so higher than they are now. Vegetation change continued until about 4,000 years ago, in tune with climatic changes that saw rainfall and temperatures continuing to fluctuate, sometimes above and sometimes below current levels.
In addition to these variations, probably the greatest long-term regional shift was the emergence and increasing strength and frequency of the bane of Australia's farmers today, El Niño-Southern Oscillation (ENSO) events. Becoming established in its current broad pattern between 7,000 and 5,000 years ago and with a peak between 3,000 and 1,000 years ago following an abrupt rise in event magnitude, ENSO significantly increases climatic variability and, in particular, can cause prolonged periods of severe drought throughout Oceania and especially in the west.[v]
AS THE EARLIER description of life at Lake Mungo shows, archaeologists have built up a solid outline of life in Australia during the last glacial, which came at the end of a geological epoch known as the Pleistocene. New data have produced a major change in perspectives on Pleistocene culture. Until recently, Ice Age life was portrayed as simple and static, in complete contrast with the more dynamic behaviour that supposedly emerged in postglacial times, known as the Holocene. Previous models painted the process as one of economic, social and political "intensification", or increased complexity, especially over the past 5,000 years. We now know that while cultural change never stops, the behaviour of modern humans, such as those who originally colonised Australia 50,000 years ago, is always complex. Indeed, such complexity is the marker of modern humanity in an anthropological/archaeological sense. Without it, people could not have colonised the continent and nearby islands in the first place. This is evident from the absence of pre-modern humans such as Homo erectus, which made it to what at the time was the tip of the Asian mainland (now Java), but no further[vi]. This cultural complexity also facilitated people's adjustment to the extraordinary environmental changes that confronted them over the millennia following colonisation.
So what happened? People had spread across all of Sahul and the Bismarck and probably the Solomon Islands archipelagos by the time the environment began to deteriorate dramatically 35,000 years ago. What they did as the Ice Age proper descended upon them depended significantly on where they were. Three examples should be enough to highlight the sort of variation: the desert, south-west Tasmania and the islands of the Bismarck Archipelago.
In the expanding arid and semi-arid zones, people contracted towards better-watered refuges around the coastline and in upland areas such as the Pilbara and the Macdonell Ranges in the interior. As a result, many sites were abandoned for millennia during the LGM, as large swathes of country simply became too dry to live in. This contraction occurred despite evidence for what Peter Veth characterises as cultures of "profound adaptability" rather than of great uniformity and conservatism as long thought[vii]. The Pleistocene archaeological record can be difficult to work with owing to preservation problems that have emerged over such a vast period. Often, all that archaeologists have to study are stone artefacts. Yet, as research summarised by Veth and Sue O'Connor indicates, variations in the abundance of such material at different times and in different places, the arrays of tool types present and changes in the sorts of stone they are made from show clearly that Pleistocene desert people were able to "reconfigure their economies, mobility patterns, territorial ranges, information-exchange systems and technological organisation" to cope with continuous climate change. Interestingly, O'Connor and Veth found that aridity had its greatest local impact towards the end of the LGM and that many areas remained uninhabitable until the early postglacial period. In fact, they think some places never returned to pre-LGM levels of vegetation cover[viii].
The Ice Age record studied by Richard Cosgrove and his colleagues in south-west Tasmania is as rich as the one in the desert is sparse. This is because many of the most ancient sites are in limestone caves that preserve large quantities of organic material to augment the assemblages of stone artefacts that can survive just about anywhere. Ice Age Tasmania was the only part of Sahul that even vaguely resembled glacial Europe or North America. Winter temperatures fell to minus fifteen degrees and summers were cool and brief. Much of the Tasmanian highlands were covered in permanent ice and snow was common in the lowlands. Tree cover was sparse. Unlike the situation in the desert, where there was widespread abandonment of sites through the LGM, many of the Tasmanian caves were used right through the glacial maximum until about 13,000 years ago, when dense rainforest spread across the region, greatly restricting people's access. As Cosgrove describes it, the sites have yielded "staggering amounts of stone tools, animal bones, hand stencil art, charcoal from cooking fires and bone implements for piercing skins. It is not uncommon to find 250,000 bones and 40,000 stone tools in less than a cubic metre of soil. The major human prey animal was Bennett's Wallaby...[which] represents about seventy per cent of the species found in the caves."
Historical analogies suggested that people would have visited the caves seasonally, in summer, when the weather was not too foul. Fascinatingly, however, detailed study of wallaby teeth indicated people used the sites in winter, precisely when the weather was worst. Cosgrove explains this pattern in terms of "optimal foraging strategies" which targeted the wallabies when they were fattest and most thickly furred and also least mobile and therefore most predictable to hunt. "It appears that the people of Ice-Age Tasmania were not mere victims of the capricious environment. They systematically planned their approach to their economy, structuring it in such a way as to take advantage of the resources available. It reflects the breadth of human behavioural flexibility and the ingenious approaches to problem-solving by Australia's early inhabitants."[ix]
So, too, do events and processes that were unfolding far to the north at the time, in the Bismarck Archipelago. Matthew Leavesley and other researchers have identified six "pulses" of activity in open-air and limestone cave sites in New Britain and New Ireland. The first pulse represents initial colonisation of an island world with an impoverished land fauna. In New Ireland, the sparse archaeological record suggests that people were organised in small, highly mobile groups that moved around to find the resources they required, though Robin Torrence's work on New Britain shows resources there were moved to people.
Environmental change linked with the LGM was limited in this tropical region. Yet, for reasons that are not clear at present, the second major pulse of activity coincided with the onset of glaciation elsewhere. This pulse saw Manus settled by at least 20,000 years ago, requiring an open-ocean journey of more than 200 kilometres, about seventy-five kilometres out of the sight of land. This makes it the longest sea journey up to this time anywhere in the world and the only known Pleistocene voyage beyond the limits of one-way island visibility. It implies capable marine craft and considerable navigational ability, even by modern standards. In other words, the colonisation of Manus entailed extensive planning and great skill in execution. At this time, people on New Ireland also began importing exotic wild mammals and stone, the latter from sources 350 kilometres away on New Britain. The second phase is followed by a gap in occupation through the LGM, which again remains mystifying in the absence of significant detectable environmental change in the region at the time. The third pulse, from 15,000 years ago, saw the re-occupation of sites throughout the region. In New Ireland, this pulse represents the greatest density of deposition of any period, suggesting regional populations had grown substantially. The final pulse occurred at about 12,000 to 10,000 years ago when all the New Ireland sites show their highest rates of cultural deposition, perhaps signalling some form of socioeconomic intensification. In New Ireland, the fourth pulse was followed by the abandonment of all cave sites from about 8,000 years ago, indicating yet another change in local ways of life[x].
There are three compelling lessons to be drawn from these sketches of Ice Age life in Australia and its near neighbourhood. First, continual and sometimes quite remarkable environmental change is normal and natural. Second, the impacts of even very large-scale environmental shifts vary between regions. The human effects of climatic and other environmental variation thus need to be thought about at regional and indeed local scales as much as an all-encompassing global one, because what happens in region "x" will not necessarily occur in region "y", even when the two are relatively close together in global terms. Finally, and perhaps most importantly in the context of today's concerns, the very clear message from the archaeology – and the archaeology of technologically simple societies at that – is that people manage the change in their environments and life goes on. It may not be life as we know it now, or even of a sort that the ancestors of the people making the change would easily recognise. To my mind, though, that is cause for optimism, not alarm. The capacity to make what can be radical changes to a way of life that may have persisted for centuries shows how adaptable and resilient people can be.
IT IS CRITICAL in this context to understand that large-scale natural environmental change did not cease at the end of the last glacial. Unlike the science fiction of Star Wars, this sort of thing does not just occur "long, long ago and far, far away". It happens right here and it never stops. Thus various things happened throughout postglacial times that would have affected human lives to greater and lesser degrees. The onset of the ENSO phenomenon was the most important of these shifts in terms of the cultural changes that occurred as people managed – and in some cases seems to have taken great advantage of – its impact on their lives. ENSO produced a significant decline in rainfall and an increase in climatic variability across much of the continent. There appears to have been a south-north trend in the onset of this drier period, which started about 4,500 to 5,000 years ago in southern Australia but between 4,000 and 3,800 years ago in the north. Rainfall increased from 2,000 years ago. Climatic variability reduced at the same time but remains a characteristic feature of Australia's environment.
ENSO is linked with major changes in Australia and may also be implicated in profoundly important historical shifts in Australia's Pacific neighbourhood. In the case of Australia, archaeologists have long sought to understand a substantial upswing in many parts of the continent in the manufacture of regularly shaped, finely made and mostly small stone implements from about 5,000 years ago (the "mid-Holocene"). The tools include so-called "Bondi points" or "backed blades", adzes and various sorts of points (as in "spear" points, though many may never have been hafted to anything).
Peter Hiscock convincingly contends that there is a causal connection between the increase in the production of these artefacts and the onset of an ENSO-dominated climate. As Australia began to experience a drier and more variable climate from around 5,000 years ago, the distribution and availability of food and other resources would have become harder to predict reliably. Excavated evidence indicates that this change occurred when population numbers in many regions were probably increasing slightly, when at least some human groups were moving into new environments and when the effects of postglacial sea-level rise were still being felt. Hiscock believes that increased production of the small artefacts was one widespread cultural shift that was made to help people manage their changing environment. He hypothesises that these tools were one component of a tool kit that helped reduce risk in acquiring resources. This is because, in technological terms, such artefacts are reliable, versatile and easy to maintain. People using this sort of technology would have had an advantage in difficult circumstances, prompting the upswing in their manufacture. Backed artefact production dropped off markedly as rainfall increased over the past 2,000 years and people began accommodating new pressures[xi].
Turning to the Pacific, Pleistocene people moved out as far as the end of the Solomon Islands chain in the same general period that they originally settled the expanded continent of Sahul. No one succeeded in getting any further into the Pacific – what archaeologists call Remote Oceania – for another 2,000 generations. This initial move beyond the Solomons about 3,000 years ago coincided with the striking increase in the frequency and intensity of ENSO events, which some scholars think assisted the colonisation process. Atholl Anderson has scrutinised existing explanations of the timing and geographic pattern of human dispersal in Remote Oceania. He proposes that the pattern of initial settlement now evident in securely dated archaeological sites demonstrates the arrival in the western Pacific of new maritime technology in the form of the sail. He thinks it was probably accompanied by paddles for steering as well as propulsion, both of which facilitated "the development of navigation with vessel controllability".
Despite the advance that such technology represented, it still only enabled travel downwind. Prevailing winds in the South Pacific are southeasterlies, which prevented travel further out into the Pacific. Seasonal reversals do occur and, in fact, were relied upon in historical times for inter-island trading. But they would have been a slow and unreliable way to move, especially over some of the longer distances involved in Polynesia. Moreover, there is clear archaeological evidence that the process of colonisation was episodic. Following earlier suggestions by other researchers, Anderson argues that anomalous westerlies created by episodic ENSO conditions would have provided the motive power for more rapid and reliable, but also punctuated, easterly migration. Tellingly for this scenario, while the major west-east movements coincided with ENSO conditions, movements in the opposite direction, for example from Melanesia into Central Micronesia, occurred during normal weather dominated by south-easterly trade winds[xii]. Plainly, ENSO was not all bad for people in the past.
In this connection, it is possible ENSO conditions also had something to do with postglacial events in Torres Strait. Bruno David, Ian McNiven and their colleagues have recently pushed back the dates for occupation in the western islands of the strait to about 8,000 years ago. At that time, the area was still part of the Australian mainland. The islands were abandoned after they were isolated by rising seas about 6,000 years ago, and saw only fleeting intermittent visitation. From 3,500 years ago, people's use of the islands increased significantly and probably involved renewed permanent settlement. David has proposed that this resettlement was connected with the same migration that took people into Remote Oceania, but careful consideration of the evidence to hand suggests that this is very unlikely. However, it seems likely that there is a broad causal link in the form of the intensification of ENSO conditions that coincided with both events[xiii].
LET ME DRAW things to a close by reiterating a couple of points. First, casting an archaeological eye over tens of thousands of years of human history makes it clear that people get through – and sometimes can even prosper from – major environmental change. Second, it is also evident that global patterns of climate change and their effects on people's lives have to be thought about at regional and local levels. One particularly important issue in this connection, stressed repeatedly in the climate projections of the greenhouse-focused Intergovernmental Panel on Climate Change (IPCC), as well as in the scientific literature about past environmental variation, is the ambiguity or absence of links between the northern and southern hemispheres. This is because the two have largely separate oceanic and atmospheric circulation systems. Just because millennial-scale climate change buried Europe and Canada under three kilometres of ice did not mean it did the same thing in Australia. Virtually all of the modelling of future climate is based on northern hemisphere data, so we really have no idea what these projections mean for the southern hemisphere. As the IPCC stated in 2001: "The scarce data from the southern hemisphere suggest temperature changes in past centuries markedly different from those in the northern hemisphere, the only obvious similarity being the strong warming during the twentieth century."
It is not even that straightforward. In the IPCC report's executive summary, there is the rider: "A few areas of the globe have not warmed in recent decades, mainly over some parts of the southern hemisphere oceans and parts of Antarctica. No significant trends of Antarctic sea-ice extent are apparent since 1978, the period of reliable satellite measurements." [xiv]
Whether this remains true is now in doubt. New ice core data indicate that Antarctic sea ice may have been shrinking since the 1950s. Other studies show Antarctic ice shelves are thinning, perhaps owing to as-yet undemonstrated rising sea temperatures, while others report that Antarctic glaciers have been retreating since 1945. The glaciers were advancing before that, suggesting that the current retreat may be part of a long-term natural cycle. Overall, however, a recent overview in Science shows that the reductions that are occurring are more or less in balance with ice thickening observable in other parts of Antarctica. Part of the problem of interpreting results such as these is that different tests monitor different chronological scales of environmental change. This makes it hard to determine whether observable shorter-term variation is in fact just one element of a much longer-term natural pattern[xv].
Issues of scale also apply to the melting of the Greenland ice sheet, which it is said will cause a terrifying seven metre rise in sea levels around the world[xvi]. We need to understand several things here. One is that the process is not like pouring a bucket of water into a full bathtub. The earth's crust is deformable. The extra weight of melt water will push the sea bed down, in turn forcing continental margins up. This will vary locally, affecting the amount the sea rises from place to place. Greenland as a whole will probably rise as well, without the weight of ice pressing down on it.
We need to appreciate is that completely melting the ice sheet would require a rise in global average temperature of about eight degrees. Since the Industrial Revolution global average temperature has risen by less than one degree, so there is some way to go. Noticeable melting will begin with less than two degrees global rise. In any case, it would take 1,000 years for half the ice sheet to melt, and 3,000 years for all of it to go[xvii]. Think about those figures for a moment. They might be just a flick of the eye in geological time, but they represent between fifty and 150 generations at a human scale. At that rate, we probably have just enough time to see what's coming and do something about it. People get unnecessarily alarmed about these sorts of projections because they confuse human and geological time scales.
To add to this uncertainty, there are also major differences within the hemispheres as well as unexpected correspondences between them that should be taken into account. A good way to emphasise this and its implications for the future is to think of past environmental change in Australia in the context of two global "slices" cutting through the continent. The first runs north-south from pole to pole and the other runs east-west around the world along the Tropic of Capricorn. The first, called the "PEP-II transect", takes in east Asia, including Japan, south and South-East Asia and Australasia. The Capricorn transect encompasses Australia, South Africa and South America. Like Australia in the southern hemisphere, but starkly unlike the rest of the northern hemisphere, the immense Tien Shan and Tibetan Plateau in western China remained largely unglaciated throughout the LGM because it was so dry (and so had no water to be made into ice). The technical explanation is that the PEP-II transect was a long way from major continental icesheets other than Antarctica and so although the "global thermal signal" registered there during the last glacial, it was mediated by local circumstances. In most parts of the transect, in fact, the global signal was overwhelmed by local rainfall patterns. This meant that while places like Japan and southern New Zealand were glaciated, others were not[xviii].
Turning to the Capricorn transect, the distinctiveness of the Tien Shan and Tibetan Plateau in the northern hemisphere during the LGM is mirrored by the distinctiveness of South America's Andes in the southern hemisphere. Like Australia, the LGM in South Africa was cold and very arid. In South America, however, now-arid parts of the west coast were wet throughout the height of the glacial. Again, it was a matter of local rainfall regimes: the Andes were watered by summer monsoons drawing moisture from the Amazon basin[xix]. This is much the same reason that the South Island of New Zealand had glaciers when Australia did not: there was much more moisture to freeze. All of this means that regardless of how accurate current modelling of future climate might be, its specifics are primarily concerned with the northern hemisphere – and Europe and North America in particular – and probably do not give us much idea of what might happen in the southern hemisphere in general or Australia in particular.
In his recent bestseller Collapse (Penguin, 2005), Jared Diamond examines a range of historical evidence concerning the success or failure of various past societies based on what he understands to have been their different approaches to environmental management. At the end of the book, he states that his "remaining cause of hope" for the future is that the modern world has archaeologists to find out what happened in the past, and TV to tell everyone else about it so they can deal with change accordingly"[xx].
The story I have told you here may not make it to television, but who am I to argue with a figure of Diamond's stature about the value of archaeological knowledge in assessing our options for the future? Like him, I think humanity's past experience gives us good reason to be prudent. Just like the people in Australia's deserts in the last glacial maximum, we should control the things we can control so they don't worsen the impact of things we can't control. Yet, on the basis of what I know about the past, I remain optimistic that people will manage. In the end, the best advice this archaeologist has to offer comes from that great futurologist Douglas Adams: DON'T PANIC!
[i] D.J. Mulvaney and J. Kamminga, 1999, Prehistory of Australia, Sydney: Allen & Unwin, p197-8.
[ii] The best recent survey of past environmental change in Australia and surrounding regions is J. Dodson et al. (eds), 2004, "Climates, human, and natural systems of the PEPII transect", special edition of Quaternary International, vol 118-119.
[iii] Mulvaney and Kamminga, p120.
[iv] S. Haberle and B. David, 2004, "Climates of change: human dimensions of Holocene environmental change in low latitudes of the PEPII transect", Quaternary International, vol 118-119:165-79, pp166-167, 176; M. Gagan et al., 2004, "Post-Glacial Evolution of the Indo-Pacific Warm Pool and El Niño-Southern Oscillation", Quaternary International, 118-119:127-43, pp131-2; though cf. Holdaway et al., 2002, "Variability in the Chronology of Late Holocene Aboriginal Occupation of the Arid Margin of Southeastern Australia", Journal of Archaeological Science 29:351-63.
[v] See Gagan et al. 2004 for technical details.
[vi] The position of the so-called "hobbits" on Flores in this connection has yet to be finalised, but it can be noted that they didn't make it to Australia either.
[vii] P. Veth, 2005, "Between the desert and the sea: Archaeologies of the Western Desert and Pilbara regions, Australia", in M. Smith and P. Hesse (eds) 230S. Archaeology and Environmental History of the Southern Deserts, pp132-41, Canberra: National Museum of Australia Press, p140.
[viii] S. O'Connor and P. Veth, 2006, "Revisiting the past: Changing interpretations of Pleistocene settlement, subsistence and demography in northern Australia", in I. Lilley (ed.) Archaeology of Oceania: Australia and the Pacific Islands, pp31-47. Oxford: Blackwell, pp43-4.
[ix] R. Cosgrove, 2004-2005, "Ice-Age hunters of Tasmania", Nature Australia Summer:58-63, pp60, 63.
[x] M. Leavesley, 2006, "Late Pleistocene complexities in the Bismarck Archipelago", in I. Lilley (ed.)Archaeology of Oceania: Australia and the Pacific Islands, Oxford: Blackwell, pp189-204. R. Torrence et al., 2004, "Pleistocene colonisation of the Bismarck Archipelago: New evidence from West New Britain",Archaeology in Oceania 39:101-30.
[xi] P. Hiscock, 2006, "Blunt and to the point: Changing technological strategies in Holocene Australia", in I. Lilley (ed.) Archaeology of Oceania: Australia and the Pacific Islands, Oxford: Blackwell, pp44-68; also Haberle and David, 2004.
[xii] A. Anderson, 2003, "Initial human dispersal in Remote Oceania: pattern and explanation", in C. Sand (ed.) Pacific Archaeology: Assessments and Prospects, pp.71-84. Nouméa: Département Archéologie, Service des Musées et du Patrimoine de Nouvelle-Calédonie.
[xiii] B. David et al., 2004, "Badu 15 and the Papuan-Austronesian settlement of Torres Strait", Archaeology in Oceania 39:65-78.
[xiv] J. Houghton et al. (eds), 2001, Climate Change 2001:The Scientific Basis. Contribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change: Cambridge: CUP, pp5, 90-1.
[xv] M. Curran et al. 2003 Ice core evidence for Antarctic sea ice decline since the 1950s : Science 302:1203-6; A. Shepherd et al. 2003 Larsen ice shelf has progressively thinned. Science 31:856-9; A. Cook et al. 2005 Retreating glacier fronts on the Antarctic Peninsula over the past half-century. Science 308:541-4; R. Alley et al. 2005 Review. Ice sheet and sea level changes. Science 310:456-60.
[xvi] L. Dayton 2006 Island nations at risk from melting glaciers. The Weekend Australian February 18-19:6.
[xvii] UK Government 2005 Stabilising climate to avoid dangerous climate change – a summary of relevant research at the Hadley Centre. Department of Environment, Food and Rural Affairs Met Office p. 8); Australian Government 2005 Climate Change Risk and Vulnerability. Promoting an efficient adaptation response in Australia. Australian Greenhouse Office, Department of the Environment and Heritage p.38.
[xviii] J. Dodson et al., 2004, "Climate, human, and natural systems of the PEP II transect", Quaternary International 118-119: 3-12; Y. Ono et al., 2004, "Timings and causes of glacial advances across the PEP-II transect (East-Asia to Antarctica) during the last glaciation cycle", Quaternary International 118-119 55-68, pp59, 65-6.
[xix] D. Thomas, 2005, "Late Quaternary environmental history of the southern deserts", in M. Smith and P. Hesse (eds) 230S. Archaeology and Environmental History of the Southern Deserts, pp14-28. Canberra: National Museum of Australia Press.
[xx] J. Diamond, 2005, Collapse: How Societies Choose to Fail or Survive, London: Penguin, p525.