|
Article Excerpt
Abstract
This paper explores the subsistence records from cave sites with Pleistocene-aged deposits in East Timor and the Aru Islands during the Last Glacial Maximum (LGM), and discusses these records within the context of the limited archaeological evidence for LGM occupation from elsewhere in the Indonesian Archipelago. Although Timor and the Aru Islands are at similar latitudes, the onset of aridity had markedly different impacts on the settlement and subsistence choices available to hunter-gatherers in these two regions. We suggest that the different occupation trajectories seen in Timor and Aru during the LGM are related, at least in part, to biogeographic contrasts across the Indonesian Archipelago.
Keywords: Last Glacial Maximum (LGM), East Timor, Aru Islands, eastern Indonesia, Pleistocene fauna.
**********
[FIGURE 1 OMITTED]
Recent research in East Timor and the Aru Islands in eastern Indonesia (Figure 1) has brought to light some significant differences in the occupation record of these islands, with the most pronounced contrast relating to the Last Glacial Maximum (LGM). Several sites with long sequences beginning between ca. 28,000 calBP and 42,000 calBP have now been excavated and analysed from East Timor and Aru. Although at roughly similar latitudes the onset of aridity appears to have had very different impacts on hunter-gatherers in these two regions. These different histories during the LGM are probably related at least in part to biogeographic differences across the Indonesian Archipelago. Unlike the continental faunas of Sundaland and Sahuland, the oceanic faunas of Wallacea (Sulawesi, the Moluccas and Lesser Sunda Islands, including Timor) are 'unbalanced' (sensu van den Bergh et al. 2001), reflecting their isolation from mainland source areas and their relatively small landmass areas. In contrast, the contemporary fauna of Aru is more diverse and 'balanced', reflecting a closer connection with Sahul until the terminal Pleistocene, and the late Pleistocene fauna was richer again (Aplin and Pasveer 2005). This paper explores the evidence of Pleistocene colonisation and subsistence behaviour for East Timor and the Aru Islands, and compares these records with the more limited archaeological evidence for LGM occupation from elsewhere in the Indonesian Archipelago.
LGM Palaeoecology and Palaeoclimate
Around 30,000 yrs ago global climates and environments experienced a dramatic shift, marked most conspicuously by a rapid buildup of glacial ice in the northern hemisphere and by an associated, sharp fall in global sea level (Lambeck and Chappell 2001:680). This paper is primarily concerned with the effects of decreased sea levels and climate change on human settlement patterns and subsistence in the Indonesian Archipelago during the LGM buildup from 30,000 yrs ago to 25,000 yrs ago and the peak LGM from approximately 25,000 yrs ago until about 15,000 yrs ago. LGM conditions, which correspond approximately with OIS2 in the deep sea core records, persisted with little respite for around 15,000 years. In most areas of the globe it had a dramatic impact on plants and animals and human populations dependent on them for survival. The northern hemisphere story is well known, where advancing ice sheets led to profound changes in vegetation and faunal distributions and rendered many areas unsuitable for human habitation (Gamble et al. 2004). In parts of the southern hemisphere environmental transformations were just as dramatic, with large areas of inland and southern Australia covered with mobile sand desert and cold, dry steppe (Hiscock and Wallis 2005; Hope et al. 2004). High altitude parts of the southern Alps and Tasmania were also ice covered during the height of this phase, which peaked approximately 18,000 yrs BP (or 21,000 calBP; Hope et al. 2004).
Much less is known however about the impact of last glacial phase climates on the tropical equatorial realms.
Until recently the evidence for environmental change in the tropics during the Quaternary was scattered, and the obvious absence of former ice sheets led to the suggestion that the tropical regions had evaded the fluctuations of temperate areas, acting as a kind of planetary refuge (Hope 2005:25).
It is now clear that this region did not go unaffected. The most abundant and striking evidence for change comes from the larger mountain ranges, where montane forest zones shifted to lower elevations and were replaced by forest types normally found at higher elevations. Above 2800 m or so, large areas were occupied by subalpine grasslands and shrublands (Hope 2005:29) while smaller areas, generally thought to be at above 3800 m, were covered by snow or glacial icecaps (Flenley and Morley 1978; Hope 2005; Peterson et al. 2001). Recent reassessments of the snowline data for Papua New Guinea and Mount Jaya in Papua at 5000+ m indicate that snowlines were at higher altitude than previously inferred from pollen data. This reassessment has major implications for earlier inferences of cooling--a snowline of 3450 m implies a reduction in mean annual temperatures of 6-8[degrees]C for LGM highland New Guinea (Hope 2001; Peterson et al. 2001), whereas current snowline projections of 4000 m for the LGM could be achieved with a mean temperature decrease of 2.5-3.0[degrees]C (Prentice et al. 2005).
Lowland tropical regions are poorly represented in terms of palaeoecological proxies during this period. However, available evidence suggests that the magnitude and direction of change was regionally variable. This is not surprising, given the great variability in annual rainfall that exists today over short distances and the extent and duration of seasonality. This variability is a product of complex relationships between landmass, altitude, wind circulation patterns and sea surface temperature and these same factors make it difficult to model environmental changes in the past, especially on the smaller islands of eastern Indonesia. Among the few proxy records from the tropical lowlands, the long pollen core from Lake Hordorli, Papua, at 780 m, is said to show "a former depression of lower montane forest next to the equatorial Pacific" indicating that late Pleistocene climates were sufficiently cool and cloudy to cause long lasting changes to forest composition, even at this altitude (Hope 2001:143). However, the Hordorli record contains no indication for drier conditions or increasing fire frequency during this time (Hope 2001:143). Similar indications come from faunal evidence from archaeological sites at 350 m on the Bird's Head of Papua, with montane forest mammals present during LGM times at an elevation that today supports lowland rainforest with its distinctive mammal species (Pasveer and Aplin 1998). In contrast, the faunal record from the two Pleistocene-aged caves at low altitude on the central Aru Island, Koobror, Liang Lemdubu and Nabulei Lisa, combine to give a clear record of increasingly drier conditions from 27,000 yrs ago until the terminal Pleistocene, with...
|
