Janet M. Wilmshurst

Dating the late prehistoric dispersal of Polynesians to New Zealand using the commensal Pacific rat

The pristine island ecosystems of East Polynesia were among the last places on Earth settled by prehistoric people, and their colonization triggered a devastating transformation. Overhunting contributed to widespread faunal extinctions and the decline of marine megafauna, fires destroyed lowland forests, and the introduction of the omnivorous Pacific rat (Rattus exulans) led to a new wave of predation on the biota. East Polynesian islands preserve exceptionally detailed records of the initial prehistoric impacts on highly vulnerable ecosystems, but nearly all such studies are clouded by persistent controversies over the timing of initial human colonization, which has resulted in proposed settlement chronologies varying from ≈200 B.C. to 1000 A.D. or younger. Such differences underpin radically divergent interpretations of human dispersal from West Polynesia and of ecological and social transformation in East Polynesia and ultimately obfuscate the timing and patterns of this process. Using New Zealand as an example, we provide a reliable approach for accurately dating initial human colonization on Pacific islands by radiocarbon dating the arrival of the Pacific rat. Radiocarbon dates on distinctive rat-gnawed seeds and rat bones show that the Pacific rat was introduced to both main islands of New Zealand ≈1280 A.D., a millennium later than previously assumed. This matches with the earliest-dated archaeological sites, human-induced faunal extinctions, and deforestation, implying there was no long period of invisibility in either the archaeological or palaeoecological records.

High-precision radiocarbon dating shows recent and rapid initial human colonization of East Polynesia.

The 15 archipelagos of East Polynesia, including New Zealand, Hawaii, and Rapa Nui, were the last habitable places on earth colonized by prehistoric humans. The timing and pattern of this colonization event has been poorly resolved, with chronologies varying by >1000 y, precluding understanding of cultural change and ecological impacts on these pristine ecosystems. In a meta-analysis of 1,434 radiocarbon dates from the region, reliable short-lived samples reveal that the colonization of East Polynesia occurred in two distinct phases: earliest in the Society Islands A.D. ∼1025–1120, four centuries later than previously assumed; then after 70–265 y, dispersal continued in one major pulse to all remaining islands A.D. ∼1190–1290. We show that previously supported longer chronologies have relied upon radiocarbon-dated materials with large sources of error, making them unsuitable for precise dating of recent events. Our empirically based and dramatically shortened chronology for the colonization of East Polynesia resolves longstanding paradoxes and offers a robust explanation for the remarkable uniformity of East Polynesian culture, human biology, and language. Models of human colonization, ecological change and historical linguistics for the region now require substantial revision.

Rapid landscape transformation in South Island, New Zealand, following initial Polynesian settlement

Humans have altered natural patterns of fire for millennia, but the impact of human-set fires is thought to have been slight in wet closed-canopy forests. In the South Island of New Zealand, Polynesians (Māori), who arrived 700–800 calibrated years (cal y) ago, and then Europeans, who settled ∼150 cal y ago, used fire as a tool for forest clearance, but the structure and environmental consequences of these fires are poorly understood. High-resolution charcoal and pollen records from 16 lakes were analyzed to reconstruct the fire and vegetation history of the last 1,000 y. Diatom, chironomid, and element concentration data were examined to identify disturbance-related limnobiotic and biogeochemical changes within burned watersheds. At most sites, several high-severity fire events occurred within the first two centuries of Māori arrival and were often accompanied by a transformation in vegetation, slope stability, and lake chemistry. Proxies of past climate suggest that human activity alone, rather than unusually dry or warm conditions, was responsible for this increased fire activity. The transformation of scrub to grassland by Europeans in the mid-19th century triggered further, sometimes severe, watershed change, through additional fires, erosion, and the introduction of nonnative plant species. Alteration of natural disturbance regimes had lasting impacts, primarily because native forests had little or no previous history of fire and little resilience to the severity of burning. Anthropogenic burning in New Zealand highlights the vulnerability of closed-canopy forests to novel disturbance regimes and suggests that similar settings may be less resilient to climate-induced changes in the future.

Forest disturbance in the central North Island, New Zealand, following the 1850 BP Taupo eruption

The 1850 BP Taupo eruption covered c. 30 000 km2 of the central North Island with airfall deposits and c. 20 000 km2 with ignimbrite. This paper reviews pollen and charcoal analyses of lake and peat sediment cores from sites at various distances and directions from the Taupo vent to establish the effects of this eruption on the surrounding forests. Forests within range of the ignimbrite were destroyed, and forests located up to 170 km east of the vent suffered variable degrees of damage from ashfall. Stands of Pteridium esculentum and other seral taxa flourished immediately after the eruption. Fires occurred during the eruption and continued for several decades after. The degree and nature of vegetation disturbance above the Taupo Tephra varied according to the thickness of ashfall, local topographical features and probably the vigour of the forest. Revegetation was complete within 200 years of the eruption, even at sites overwhelmed by the Taupo Ignimbrite. Post-eruption forests were similar to those existing before the Taupo eruption.

Late Holocene forest disturbance in Gisborne, New Zealand: A comparison of terrestrial and marine pollen records

Abstract A late Holocene (from c. 5500 yr B.P.) record of vegetation change is presented for the Gisborne region, based on pollen, charcoal, and tephra analyses of a terrestrial and a marine core. Up until the time of anthropogenic deforestation about 650 yr B.P., well drained lowland areas were covered with a Prumnopitys taxifolia, and Dacrydium cupressinum‐dominated podocarp/hardwood forest. The poorly drained Dacrycarpus dacrydioides‐dominated alluvial swamp forests were not as vulnerable to fire, and remained on the Gisborne Plains until European drainage and clearance began in the 19th century. In the last 5500 yr B.P., the lowland forests have been disturbed by at least five ashfalls originating from volcanic eruptions in the Central Volcanic Region. Where the terrestrial and marine cores overlap, comparisons of the pollen records show the vegetation changes and taxa present to be comparable. The fire record was not clear in the marine record, as the charcoal curve was diluted with high background l...

Asynchronous climate change between New Zealand and the North Atlantic during the last deglaciation

Climatic fluctuations recorded in Antarctica and Greenland during the last deglaciation (18–10 ka) differ markedly in their timing. It remains controversial whether local climate fluctuations recorded in southern mid-latitudes relate primarily to northern or southern polar records. We present multiproxy results from New Zealand that show strong evidence for a minor cooling or slowdown in the rate of warming at the time of the North Atlantic late glacial interstadial. The Younger Dryas chronozone in New Zealand was a period of resumed warming and increased westerly airflow. Differences between the hemispheres at this time were probably due to a reorganization of the thermohaline circulation system and associated changes in the meridional temperature gradient.

Using rat-gnawed seeds to independently date the arrival of Pacific rats and humans in New Zealand

The Pacific rat (Rattus exulans) was transported throughout the south Pacific with voyaging humans. Thus, the earliest dated evidence of Pacific rat can be used to infer first human contact. Until recently, it was considered that rats arrived in New Zealand with humans in the thirteenth century AD. However, controversial radiocarbon dates on Pacific rat bones now suggest that rats reached the remote islands of New Zealand with people c. AD 50-150. These dates are anomalous because they imply human contact with New Zealand more than 1000 years before any archaeological evidence for human presence, and precede settlement of tropical eastern Polynesia, the ancestral homeland of Maori, the first New Zealanders. The early rat bone dates are controversial for other technical reasons, which have been debated in the literature. Here, distinctive rat-gnawed seed cases preserved in sediments are used as a proxy to independently date the arrival of the Pacific rat and humans in New Zealand. This method effectively bypasses the problems that have plagued rat bone dating and provides a reliable age for rat and human arrival. The oldest dates on rat-gnawed seed cases from widely separated sites are consistent with the Pacific rat arriving at the same time as the initial human settlement of New Zealand in the thirteenth century AD, and not before. The gnawed seed dates lend no support to the argument for an earlier introduction of rats. This dating approach offers a novel way of clarifying island colonization histories throughout Oceania.

Rapid deforestation of South Island, New Zealand, by early Polynesian fires

In most parts of the world where people have colonized and modified their landscapes for several millennia or more, it is often difficult to discriminate anthropogenic burning from natural fire regimes that are linked to climate regimes. New Zealand provides a unique setting for identifying human influence on fire occurrence because it was settled recently (c. AD 1280) at a time when climates are considered to be similar to today. Late-Holocene pollen and charcoal records from New Zealand provide striking evidence for initial Polynesian (Māori) arrival being strongly associated with widespread burning and loss of native forest. The duration of initial forest clearance and the spatial pattern of burning that led to this transformation are still poorly understood. We present high-resolution charcoal and pollen analyses of sediment cores from five lakes, located on the deforested eastern side of the Southern Alps. These records document the local fire history of the last 1000 years and the response of vegetation and watersheds to burning. Our results suggest that one to several high-severity fires occurred within a few decades of initial Māori arrival, and this ‘Initial Burning Period’ (IBP) resulted in the majority of forest loss and erosion. Changes in sedimentation rates, soil chemistry and magnetic susceptibility occurred simultaneously with the first fires at some sites, and marked the end of the IBP at others, suggesting substantial and rapid alteration of watershed vegetation, soil and biochemistry. Timing of the beginning of the IBP varied across sites but the duration of this period was brief (decades to a century). Our results suggest that Māori burning of native forests was deliberate and systematic. These forests had no previous history of fire and thus showed little resilience to the introduction of a new disturbance.

A Holocene record of climate, vegetation change and peat bog development, east Otago, South Island, New Zealand

A Holocene record of pollen, macrofossils, testate amoebae and peat humification is presented from a small montane bog. Sediment accumulation began before 9000 yr BP, but peat growth not until ca. 7000 BP. From 12 000 to 7000 yr BP, a shrub-grassland dominated under a dry climate, with increasing conifer forest and tall scrub from ca. 9600 yr BP. At 7000 yr BP a dense montane-subalpine low conifer forest established under a moist, cool climatic regime. Between 7000 and 700 yr BP the bog surface was shrubby, tending to be dry but with highly variable surface wetness. The catchment was affected by major fire at least four times between 4000 and 1000 yr BP. Both fire and bog surface wetness may have been linked to ENSO-caused variations in rainfall. Cooler, cloudier winters and disturbance by fire promoted the expansion of the broadleaf tree Nothofagus menziesii between 4000 yr BP and 1300 yr BP at the expense of the previous conifer forest-scrub vegetation. Polynesian fires (ca. 700 yr BP) reduced the vegetation to tussock grassland and bracken. Deforestation did not markedly affect the hydrology of the site. European pastoralism since ad 1860 has increased run-off and rising water tables in the bog have led to a Sphagnum-dominated cover. Copyright

Contribution of floodplain sequestration to the sediment budget of the Waipaoa River, New Zealand

Abstract Rapid vertical accretion on the Waipaoa River floodplain is conditioned by the river’s high suspended sediment load (30 000–40 000 mg l−1 at flood stage). Cumulative sediment accumulation curves derived from three cores suggest an average (post-1850) rate of vertical accretion of c. 60 mm a−1, though a 15 year lacuna in flood activity has depressed the post-1948 rate to c. 40 mm a−1. Rates of aggradation during floods are several orders of magnitude larger than the time-averaged rate. Within a 44 km long reach, cross-section surveys indicate that 0.2–0.8 m of sediment was deposited between 1979 and 1990. Over this period floodplain storage accounted for 5% of the total suspended sediment load, and 16% of the suspended sediment load transported during events that exceeded bankfull stage. The Waipaoa River floodplain may be representative of floodplains bordering rivers with high suspended sediment loads, produced by rapid, episodic vertical accretion, on which overbank deposition occurs across the entire floodplain, and is complemented by channel aggradation. Such rivers are able to construct high banks. Thus channel capacities are greater and the incidence of overbank flows is less than in rivers where overbank deposition is slow relative to the rate of floodplain destruction by lateral migration. The difference between our time-averaged estimate for sequestration on the Waipaoa River floodplain and comparable estimates for actively meandering rivers, and meandering rivers with low sediment loads, reinforces the notion that there is a link between the sediment transport regime of a river and its sedimentary record. To elucidate this link it is necessary to view vertical accretion in the context of the flood events that generated it, rather than in the context of a time-averaged sediment budget.