Matthew Prebble

Prehistoric human impacts on Rapa, French Polynesia

New excavations and survey on the island of Rapa have shown that a rockshelter was occupied by early settlers around AD 1200 and the first hill forts were erected about 300 years later. Refortification occurred up to the contact period and proliferated around AD 1700. Taro cultivation in terraced pond-fields kept pace with the construction of forts. The authors make a connection between fort-building and making pond-fields, demonstrating that the pressure on resources provoked both the intensification of agriculture and hostility between the communities of the small island.

Detecting the initial impact of humans and introduced species on island environments in Remote Oceania using palaeoecology

The isolated archipelagos of Remote Oceania provide useful microcosms for understanding the impacts of initial human colonization. Palaeoecological data from most islands reveal catastrophic transformations, with losses of many species through over-hunting, deforestation and the introduction of novel mammalian predators, the most ubiquitous and devastating being commensal rats. Two case studies from the Austral Islands and New Zealand demonstrate the potential of direct human proxies from palaeoecological archives to detect initial human impacts on islands. We show how pollen from introduced crop plants, and buried seeds with gnaw marks from the introduced Pacific rat (Rattus exulans) provide a reliable means of detecting initial human colonization and highlight the downstream ecological consequences of agriculture and rat introduction on previously uninhabited pristine island ecosystems. Previous studies have relied on indirect signals of human arrival based on charcoal and associated vegetation changes, the causes of which are often more difficult to interpret with certainty.

Archaeobotany in Australia and New Guinea: practice, potential and prospects

Abstract Archaeobotany is the study of plant remains from archaeological contexts. Despite Australasian research being at the forefront of several methodological innovations over the last three decades, archaeobotany is now a relatively peripheral concern to most archaeological projects in Australia and New Guinea. In this paper, many practicing archaeobotanists working in these regions argue for a more central role for archaeobotany in standard archaeological practice. An overview of archaeobotanical techniques and applications is presented, the potential for archaeobotany to address key historical research questions is indicated, and initiatives designed to promote archaeobotany and improve current practices are outlined.

Forest clearance and agricultural expansion on Rapa, Austral Archipelago, French Polynesia

Palynological records from Holocene wetland deposits in East Polynesia have demonstrated widespread ecological changes following Polynesian arrival after c. ad 1200, but linking inferences of human activities to archaeological records has been limited by equivocal fossil proxies and a lack of chronological controls. To address these limitations, multiple sedimentary profiles were examined from a coastal marsh on the remote East Polynesian island of Rapa. These profiles span 8000 years of ecological change and record mid-Holocene sea-level highstand conditions which receded to modern levels by ad 500. Depositional models were constructed for each profile using Bayesian inferences to characterise the spatial and temporal changes in fossil proxy representation. Just prior to human arrival there are high pollen concentrations of Pandanus and the presence of an extinct palm, both indicative of an extensive lowland swamp forest that developed after ad 500. Indicators of Polynesian arrival and agricultural expansion include unprecedented amounts of charcoal particles and pollen from the introduced cultigen, Colocasia esculenta (taro). The swamp forest was progressively cleared, beginning in the most inland section at ad 1110–1230. The Colocasia-based agricultural system reached its greatest extent from ad 1590 to 1740. The last appearance of the extinct palm was recorded at ad 1520–1660 and the entire area was cleared of indigenous trees soon after European colonisation at ad 1830–1880. By modelling the chronologies of individual fossil proxies across each profile, we have developed a framework for defining the processes behind vegetation change and for matching palynological-based inferences of human activity with archaeological records.