Southern Cook Islands Research Articles

High-resolution reconstruction of infiltration in the Southern Cook Islands based on trace elements in speleothems

AbstractThis study utilizes speleothem trace elements as climate proxies to reconstruct hydroclimate variability over approximately 350 years in the Southern Cook Islands. Stalagmites Pu17 and Pu4 from Pouatea cave were analyzed using high-resolution LA-ICP-MS for trace elements (Mg, Na, Sr, P, U, Y). By monitoring cave dripwater and conducting regression analysis, we found that Mg, Sr, and Na in Pouatea dripwater mostly originated from marine aerosols, while Sr and Ba were primarily from bedrock, with additional Ba coming from marine aerosols and weathered oceanic basalt leaching. Mg was identified as the most reliable element for hydroclimate reconstruction due to its predominantly marine aerosol origin. Infiltration, via dilution of marine aerosols and bedrock inputs, was identified as the main driver of trace element variations in Pouatea at a seasonal scale. Transfer functions were established between each trace element and effective infiltration was calculated, with Mg showing the strongest correlation. The reconstructed infiltration data were compared with climate indices, showing an overarching role of the SPCZ and ENSO in controlling rainfall in the South Pacific. This research demonstrates the potential of speleothem trace elements for paleohydroclimate reconstructions, improving understanding of rainfall variability in the climatically vulnerable South Pacific Islands over the past millennia.

Neogene and Quaternary reef terraces of Mangaia, Rarotonga and Aitutaki (southern Cook Islands, South Pacific) revisited: gauges of sea-level change and island uplift?

New sedimentological data of facies and diagenesis as well as chronological data including strontium (87Sr/86Sr)-isotope ratios and uranium (U)-series dating, radiocarbon (14C) accelerator mass spectrometry (AMS) dating and biostratigraphy from elevated reef terraces (makatea) in the southern Cook Islands of Mangaia, Rarotonga and Aitutaki contribute to controversial discussions regarding age and sea-level relationships of these occurrences during the Neogene and Quaternary. The oldest limestones of the uplifted makatea island of Mangaia include reef-related facies which are mid-Miocene in age, based on new Sr-isotope and biostratigraphical data. In between these older deposits and the lowest coastal reef terrace of marine isotope stage (MIS) 5e, various older Pleistocene reef-related facies were identified. Based on Sr-isotope ratios, these were deposited during earlier Pleistocene highstands (as old as 2.28 Ma). Rare reef terraces on Rarotonga belong to the Plio-Pleistocene and the late Miocene, according to 87Sr/86Sr ratios. The late Miocene age is enigmatic as it exceeds the age of subaerially exposed volcanic rocks of Rarotonga island. The fossil reef could have formed on an older submarine volcanic high that was later displaced by younger volcanism to its present position, or the Sr-age could be too old due to diagenetic resetting. The Plio-Pleistocene Rarotonga reef terraces are overlain irregularly by Holocene reef deposits that are interpreted as storm rubble. Reef terraces on Aitutaki represent evidence of a higher-than-present (up to 1 m) sea-level during the late Holocene, based on 14C AMS age data. They are very similar to elevated late Holocene reefs of adjacent French Polynesia with regard to composition, elevation and age.

Controls on rainfall variability in the tropical South Pacific for the last 350 years reconstructed from oxygen isotopes in stalagmites from the Cook Islands

The hydroclimate of the South Pacific is highly variable on a range of timescales, however, the brevity of instrumental data in the region complicates the detection and attribution of this variability. Annually laminated speleothems may fill this knowledge gap by providing paleo-hydroclimate proxy data extending beyond the instrumental period. In this study, replicated oxygen isotope (δ18O) time series from two stalagmites from Pouatea (Pu17) and Nurau (Nu16) caves in the southern Cook Islands were used to reconstruct the variability of effective infiltration over the past 350 years (1672–2019 CE) based on observational rainfall data and cave monitoring and calcite farming experiments. The reconstructed infiltration was then compared to available reconstructions of known climate drivers of the South Pacific. Significant dry and wet periods over the last 350 years were identified based on the δ18O signal of compact/porous calcite fabric and their δ13C signals. The hydroclimate significance of these proxies was validated through comparison with the instrumental records. We could not reach an annual resolution of reconstructed infiltration, due to a combination of stable isotope sampling limitations and mixing effects within the aquifer. Nevertheless, the combined δ18O record of Pu17 and Nu16 identifies that the main driver of infiltration and, hence rainfall variability, in the southern Cook Islands is the location of the South Pacific Convergence Zone (SPCZ), which in turn is modulated by the El Niño Southern Oscillation (ENSO). This study demonstrates the hydroclimate significance of speleothem fabrics when combined with O and C stable isotopes, and adds paleoclimate information to a region that has a paucity of data.

Interdecadal modulation of the effect of ENSO on rainfall in the southwestern Pacific

The El Niño-Southern Oscillation (ENSO) is the dominant driver of interannual variability on rainfall in many Pacific Islands and in countries bordering the tropical Pacific Ocean. From 1916 through to 1975, the correlation coefficient between the Southern Oscillation Index (SOI) and interannual variability in rainfall in eastern Australia was strong in negative phases of the Interdecadal Pacific Oscillation (IPO) but weak in positive phases. By examining records of rainfall over the past hundred years in central Vanuatu and on the ‘dry side’ of Fiji, which both lie near the southern edge of the South Pacific Convergence Zone (SPCZ), we find that such modulation by IPO has been much weaker there than in eastern Australia. This paper examines possible reasons for this difference. We also find that the correlation between rainfall and the SOI remained strong throughout each of the past three phases of the IPO, in all these places, including eastern Australia. However, at Rarotonga in the southern Cook Islands, whose position is also near the southern edge of the SPCZ, but at the southeastern end, the displacement of the SPCZ by ENSO events is greater there than further west. Consequently, the correlation between rainfall and SOI is so strong at Rarotonga in El Niño years with SOI < −5 that SOI alone becomes a good predictor of wet-season rainfall there. The difference in modulation of rainfall in eastern Australia between the two positive phases of IPO (1926–1941 and 1978–1998) may be due to the influence on Australia of other climatic oscillations, such as the Indian Ocean Dipole, though other factors may also have played a role.

Human settlement of East Polynesia earlier, incremental, and coincident with prolonged South Pacific drought

The timing of human colonization of East Polynesia, a vast area lying between Hawai'i, Rapa Nui, and New Zealand, is much debated and the underlying causes of this great migration have been enigmatic. Our study generates evidence for human dispersal into eastern Polynesia from islands to the west from around AD 900 and contemporaneous paleoclimate data from the likely source region. Lake cores from Atiu, Southern Cook Islands (SCIs) register evidence of pig and/or human occupation on a virgin landscape at this time, followed by changes in lake carbon around AD 1000 and significant anthropogenic disturbance from c. AD 1100. The broader paleoclimate context of these early voyages of exploration are derived from the Atiu lake core and complemented by additional lake cores from Samoa (directly west) and Vanuatu (southwest) and published hydroclimate proxies from the Society Islands (northeast) and Kiribati (north). Algal lipid and leaf wax biomarkers allow for comparisons of changing hydroclimate conditions across the region before, during, and after human arrival in the SCIs. The evidence indicates a prolonged drought in the likely western source region for these colonists, lasting c. 200 to 400 y, contemporaneous with the phasing of human dispersal into the Pacific. We propose that drying climate, coupled with documented social pressures and societal developments, instigated initial eastward exploration, resulting in SCI landfall(s) and return voyaging, with colonization a century or two later. This incremental settlement process likely involved the accumulation of critical maritime knowledge over several generations.

Ancient DNA evidence for the introduction and dispersal of dogs in New Zealand

When people first arrived in New Zealand around 700 years ago, they brought their dogs (Canis familiaris) with them. To investigate the introduction and dispersal of dogs across the country we generated twenty-three new complete, or nearly complete, mitogenomes from ancient DNA from dog teeth sampled from four early archaeological sites in New Zealand and from one archaeological site in the southern Cook Islands. When considered together with fourteen previously reported mitogenomes from the New Zealand colonisation era site of Wairau Bar these sequences reveal a striking lack of mitochondrial genetic diversity in early New Zealand dogs. Our analysis shows that a group of closely-related dogs were brought to New Zealand, probably from an East Polynesian source population, and that these dogs and their offspring were widely dispersed throughout the country during the colonisation process. This pattern is consistent with the current model of rapid colonisation of New Zealand undertaken by highly mobile groups of people.

Cook Island artifact geochemistry demonstrates spatial and temporal extent of pre-European interarchipelago voyaging in East Polynesia

The Cook Islands are considered the "gateway" for human colonization of East Polynesia, the final chapter of Oceanic settlement and the last major region occupied on Earth. Indeed, East Polynesia witnessed the culmination of the greatest maritime migration in human history. Perennial debates have critiqued whether Oceanic settlement was purposeful or accidental, the timing and pathways of colonization, and the nature and extent of postcolonization voyaging-essential for small founding groups securing a lifeline between parent and daughter communities. Centering on the well-dated Tangatatau rockshelter, Mangaia, Southern Cook Islands, we charted the temporal duration and geographic spread of exotic stone adze materials-essential woodworking tools found throughout Polynesia- imported for more than 300 y beginning in the early AD 1300s. Using a technique requiring only 200 mg of sample for the geochemical analysis of trace elements and isotopes of fine-grained basalt adzes, we assigned all artifacts to an island or archipelago of origin. Adze material was identified from the chiefly complex on the Austral Islands, from the major adze quarry complex on Tutuila (Samoa), and from the Marquesas Islands more than 2,400 km distant. This interaction is the only dated example of down-the-line exchange in central East Polynesia where intermediate groups transferred commodities attesting to the interconnectedness and complexity of social relations fostered during postsettlement voyaging. For the Cook Islands, this exchange may have lasted into the 1600s, at least a century later than other East Polynesian archipelagos, suggesting that interarchipelago interaction contributed to the later development of social hierarchies.

Rupture and renewal in the Polynesian iconoclasm Sissons, Jeffrey (2014) The Polynesian Iconoclasm: Religious Revolution and the Seasonality of Power. New York and Oxford: Berghahn Books, 160 pp, USD85.00, hbk, ISBN: 9781782384137.

Rupture and renewal in the Polynesian iconoclasm Sissons, Jeffrey (2014) The Polynesian Iconoclasm: Religious Revolution and the Seasonality of Power. New York and Oxford: Berghahn Books, 160 pp, USD85.00, hbk, ISBN: 9781782384137.

Measuring stone artefact transport: the experimental demonstration and pilot application of a new method to a prehistoric adze workshop, southern Cook Islands

This paper presents a new method, based on the calculation of a Cortex Ratio, capable of contributing towards measures of stone artefact transport in Polynesia. A set of ratios can be calculated by comparing the observed amount of cortical surface area and volume in an assemblage with what is expected should all the products of reduction remain. Because raw material shape and size is controlled for using geometric equations, an estimate for the number of preforms produced from an assemblage also is possible. The method is experimentally demonstrated and applied to an archaeological early stage adze manufacturing assemblage from Moturakau Rockshelter, Aitutaki Island in the southern Cook Islands. Application of the method, in combination with geochemical and chronological analyses, shows that the number of preforms produced and transported, as well as the frequency of their transport, changed over time but the geographic scale of distribution remained the same, essentially local to Aitutaki.

Vegetation Changes Viewed from Pollen Analysis in Rarotonga, Southern Cook Islands, Eastern Polynesia

This study presents accelerator mass spectrometry (AMS) radiocarbon dates and pollen assemblages of 400-cm core sediments collected from the Karekare Swamp in Rarotonga, Southern Cook Islands, to investigate vegetation changes on the island, in particular those induced by human impacts. Eight 14C dates of charcoal and higher plant fragment samples indicate that the sediments accumulated since ∼6.0 cal kBP, with an apparent interruption of deposition (hiatus) from 130 to 132 cm in depth, corresponding to ∼2.8 to 0.7 cal kBP. The appearance of Chenopodiaceae pollen from upland weeds, and Cucurbitaceae and Vigna pollen grains from cultivated plants suggest that human influence existed in core sediments above 130 cm in depth. The increased abundance of Pandanus pollen and monolate-type fern spores also implies the existence of human activity.

Vegetation Changes Viewed from Pollen Analysis in Rarotonga, Southern Cook Islands, Eastern Polynesia

This study presents accelerator mass spectrometry (AMS) radiocarbon dates and pollen assemblages of 400-cm core sediments collected from the Karekare Swamp in Rarotonga, Southern Cook Islands, to investigate vegetation changes on the island, in particular those induced by human impacts. Eight14C dates of charcoal and higher plant fragment samples indicate that the sediments accumulated since ∼6.0 cal kBP, with an apparent interruption of deposition (hiatus) from 130 to 132 cm in depth, corresponding to ∼2.8 to 0.7 cal kBP. The appearance of Chenopodiaceae pollen from upland weeds, and Cucurbitaceae andVignapollen grains from cultivated plants suggest that human influence existed in core sediments above 130 cm in depth. The increased abundance ofPandanuspollen and monolate-type fern spores also implies the existence of human activity.

Short-Lived Plant Materials, Long-Lived Trees, and Polynesian 14C Dating: Considerations for 14C Sample Selection and Documentation

For over 2 decades, there have been calls for Polynesian archaeologists to identify radiocarbon samples to taxon and material type, and preferentially date short-lived materials. This stems from recognition that even modest amounts of inbuilt age are problematic in this oceanic region where human settlement dates to the last 3 millennia or less. Despite programmatic statements to this effect, and empirical demonstrations of value, uptake of these practices has been slow. This article suggests that short-lived materials be defined as those with lifespans of 10 yr or less, recognizing that reducing the distance between target event and the dated material is an ongoing aim for archaeology. We identify both short-lived materials that are suitable for dating and long-lived species (>75 yr) that should be avoided given their potential for inbuilt age. Additionally, we suggest that given the destructive nature of 14C dating there is value to (1) obtain high-resolution photographs of samples prior to analysis, (2) split samples and archive residuals when possible, and (3) routinely report the tools and expertise used in identification. Past and recent dating analyses at a Polynesian site in the southern Cook Islands are used to illustrate how the suggested protocols might enhance current and future studies. We also establish that mature Calophyllum inophyllum, a common coastal tree and valued timber species, has been in the southern Cook Islands for at least the last 700 yr and is likely to be indigenous and long-lived.

Modelling site formation dynamics: geoarchaeological, chronometric and statistical approaches to a stratified rockshelter sequence, Polynesia

Formation processes and chronology are central to many archaeological endeavours, and accuracy in the latter is dependent on control of the former. Here we consider how to integrate geoarchaeological and chronometric evidence to better understand the formational history, along with the timing and duration of cultural activities, at a regionally significant site in the southern Cook Islands (Polynesia) which dates from ∼12th century AD. Initially we model natural and cultural processes of deposition using geoarchaeological evidence. The results are integrated with 21 radiocarbon dates, including nine new AMS results on identified medium and short-lived materials. Sample outliers are identified manually and statistically and a Bayesian analysis helps to refine age estimates for the site's 12 stratigraphic zones. The timing and rate of sediment accumulation is further considered through an age–depth model. Finally, the chronology developed from these combined approaches is used to address three key processes: 1) initial site use, potentially relevant to human colonisation of this Polynesian island and archipelago; 2) on-site activities and patterns of resource use; and 3) major marine deposition events which disrupted cultural activities and may have palaeoclimatological significance. The approach developed here potentially has utility elsewhere, particularly where human activities are recurrent and of limited duration.

Subfossils of extinct and extant species of Simuliidae (Diptera) from Austral and Cook Islands (Polynesia): anthropogenic extirpation of an aquatic insect?

Subfossil head capsules of Simuliidae larvae have been recovered from swamps on Tubuai and Raivavae of the Austral Islands, and Atiu and Mangaia of the southern Cook Islands. For Tubuai and Raivavae it is likely that the simuliids are extinct, but a single simuliid species is extant on nearby Rurutu. For Atiu and Mangaia, extant simuliids have not been reported, but are known on Rarotonga. Well-preserved head capsules indicate that the Cook Islands subfossils are those of Sinulitin (Inseliellumn) teruananga Craig and Craig, 1986. For the Austral Islands, the simuliid from Tubuai is considered a variant of Simudiunt (Inseliellumn) rurutuense Craig and Joy, 2000. That from Raivavae is morphologically distinct and is described here as a new species, Simuliun (Inseliellumn) raivavaense Craig and Porch. Humans arrived in Eastern Polynesia ca. 1,000 years ago resulting in the widespread destruction of lowland forest and conversion of wetlands to agriculture with implied consequences for the indigenous biota of these habitats. Here we consider that one such result was loss of freshwater aquatic biodiversity.

“Buying in” to niche construction? A case study from the southern Cook Islands

The U–Th chronometer has revolutionized Quaternary science in the last few decades and can now be considered a well-established, mature technique with growing application in archaeology as an alternative to 14C dating and OSL, reaching well beyond the cutoff for both these methods. Applying the U–Pb technique to material within a time frame of interest to archaeology is a challenge only recently overcome, with enormous potential for future development and use. In terms of range, U–Th can generally date material of between a few years and as much as 600 ka, while U–Pb is best suited to material of 1 million years and older. U–Th dating is typically precise, with error margins of better than 1% routinely achievable for material of the last interglacial age or younger. Such precision is possible with U–Pb dating in the mid-Quaternary but most ages have uncertainties in the 5–10 % range. Carbonates are the most desirable U-series target material, with closed system behaviour being particularly important for U–Th dating; making inorganic deposits such as speleothems (secondary cave carbonates), tufa and calcrete ideal. Some biogenic carbonates such as coral and eggshell are often amenable to U-series dating, but less ideal material such as fossil teeth, bone and mollusc shell usually continue to gain uranium after deposition meaning accurate ages cannot be calculated. In all cases, the importance of a thorough understanding of the context of dated material cannot be overstated. In this contribution we review U–Th and U–Pb dating through an archaeological lens, focussing on the methods themselves; how to best interpret published data and how best to avoid common pitfalls. We highlight the major successes through a number of case studies and provide an overview of what we believe to be the future directions of this field.

The effects of natural disturbances, reef state, and herbivorous fish densities on ciguatera poisoning in Rarotonga, southern Cook Islands

Ciguatera poisoning is a critical public-health issue among Pacific island nations. Accurately predicting ciguatera outbreaks has become a priority, particularly in Rarotonga in the southern Cook Islands, which has reported the highest incidence of ciguatera poisoning globally. Since 2006, however, cases of ciguatera poisoning have declined, and in 2011 ciguatera cases were the lowest in nearly 20 years. Here we examined the relationships between cases of ciguatera poisoning, from 1994 to 2011, and: (i) coral cover, used as a proxy of reef state, (ii) the densities of herbivorous fishes, and (iii) reef disturbances. We found that coral cover was not a good predictor of cases of ciguatera poisoning, but high densities of the herbivorous fish Ctenochaetus striatus and reef disturbances were both strong predictors of ciguatera poisoning. Yet these two predictors were correlated, because the densities of C. striatus increased only after major cyclones had disturbed the reefs. Since 2006, the number of cyclones has decreased considerably in Rarotonga, because of the climatic shift toward the negative phase of the Pacific Decadal Oscillation. We suggest that fewer cyclones have led to decreases in both the densities of C. striatus and of the number of reported cases of ciguatera poisoning in Rarotonga.

Socioeconomic consequences of ciguatera poisoning in Rarotonga, southern Cook Islands

For the last 20 years, the world's highest incidence of ciguatera poisoning has been reported in Rarotonga, which is located in the southern Cook Islands. Therefore, Rarotonga is an ideal location to understand the socioeconomic consequences of ciguatera poisoning on small-island communities. In 2011, information on protein consumption preferences was collected from 179 Rarotongan households and compared with previous food consumption studies. This study showed that ciguatera poisoning halved the per-capita fresh fish consumption, from 149g/person/day in 1989 to 75g/person/day in 2006. Consequently, the consumption of alternative proteins, particularly imported meats, increased from 1989 to 2006. The cost-savings-and-avoidance valuation technique, used to estimate the direct loss in value of marketable goods and services, showed that the economic consequences of ciguatera poisoning amounted to approximately NZD $750,000 per year. Approximate costs associated with dietary shifts amounted to NZD $1 million per year. With the decline in cases of ciguatera poisoning in recent years, fresh fish has returned to the menu of residents, and the per-capita fresh fish consumption increased to 104g/person/day in 2011. Yet over the last two decades, the impact of ciguatera poisoning on the local Rarotongan community may have had long-term health-related consequences, and may have changed the social, cultural, and traditional characteristics of a once subsistence fishing lifestyle.

History as Sacrifice: The Polynesian Iconoclasm

ABSTRACTThe Polynesian iconoclasm comprised a series of destructive episodes in which god‐images were rendered powerless and temples and associated structures were successively burned or torn down in Eastern Polynesia. Beginning in Tahiti and neighbouring Mo'orea in 1815, it spread rapidly to other Society Islands, the Austral Islands and the Southern Cook Islands. In this article I suggest that the willingness of Polynesian leaders to follow Tahitian precedence in destroying or unwrapping their god images can be partially explained by the fact that they saw themselves as participants in a large‐scale sacrifice, the common objective of which was the acquisition of greatermanaand the consequent revitalisation of their societies. In making this argument I engage with anthropological theories that address relationships between sacrifice and historical transformation. I conclude by proposing that the Polynesian iconoclasm was a form of rituopraxis comprising regionally specific improvisations upon a general ritual schema.

Ciguatera poisoning in Rarotonga, southern Cook Islands

The central Pacific Ocean has arguably more ciguatera poisoning than any other place on earth. Here we tested two competing hypotheses that outline the primary causes of ciguatera outbreaks: (1) the ‘new surface hypothesis’ and (2) the ‘climate oscillation hypothesis’. Our findings indicated that in Rarotonga, from 1994 to 2010, the annual incidence of ciguatera poisoning ranged from 204 to 1,058 per 10,000 population per year. We found that the widest reefs of Rarotonga elicited the most cases of ciguatera poisoning, but found no relationship between ciguatera outbreaks and reef exposure. We also found strong correlations between cases of ciguatera poisoning and (i) the positive phase of the Pacific Decadal Oscillation, (ii) El Niño years, and (iii) periods with frequent disturbances. Yet, most disturbances occurred during the above-mentioned climate phases. This study links the two supposedly, mutually exclusive hypotheses. Moreover, as predicted by the ‘climate oscillation hypothesis’, the Pacific Ocean is now, in 2010, experiencing a negative phase of the Pacific Decadal Oscillation, and Rarotonga is reporting few cases of ciguatera poisoning.

Dynamics of Polynesian Subsistence: Insights from Archaeofauna and Stable Isotope Studies, Aitutaki, Southern Cook Islands

Human colonists of Remote Oceania readily took advantage of the naive virgin fauna encountered on previously uninhabited islands, a bounty that was quickly depleted. Subsequent developments in Polynesian subsistence economies were more subtle, varied, and complex. These features are illustrated in a comparison of two quite different subsistence archives from the postcolonization period: archaeofaunal assemblages and stable isotope (δ 13 C and δ 15 N) records of humans, pigs, and dogs from the same archaeological contexts. The samples come from four stratified sites, with a total of 22 distinct occupational strata that represent a 600-year period on the small (18.4 km 2 ) almost-atoll of Aitutaki in the southern Cook Islands. Benefits and challenges of integrating these quite different records are considered in the context of specific findings, with implications for subsistence studies elsewhere. In particular, differences in formation processes, taxonomic resolution, and contrasting spatial and temporal scales represented by each record are highlighted. A complex, multiscalar picture of subsistence change emerges, showing variability within and across the three species and the two subsistence archives. Findings support prior interpretations that established (not colonial) settlements are represented by the currently known Aitutaki archaeological record. Within the relatively stable and largely anthropogenic food web, humans occupy a central position throughout the sequence. Through time, a reduction in fishing and decreased consumption of marine carnivores is indicated; these changes are likely to be an outcome of both repeated storm events and considerable shoreline disruption in the fourteenth century A.D., and cultural decisions about the relative costs and benefits of various fishing activities vis-a-vis other subsistence needs. An apparent reduction in variability of pig diets in late prehistory could reflect interspecific competition between pigs and their human managers, although small sample sizes constrain interpretations. Overall, use of two quite different subsistence archives provides a more robust, but also more complex, view of subsistence change across individuals and communities on Aitutaki.