• 2018-07
  • 2019-04
  • 2019-05
  • 2019-06
  • 2019-07
  • 2019-08
  • 2019-09
  • 2019-10
  • 2019-11
  • 2019-12
  • 2020-01
  • 2020-02
  • 2020-03
  • 2020-04
  • 2020-05
  • 2020-06
  • 2020-07
  • 2020-08
  • 2020-09
  • 2020-10
  • 2020-11
  • 2020-12
  • 2021-01
  • 2021-02
  • 2021-03
  • Stable carbon and nitrogen isotope ratios in bone collagen a


    Stable carbon and nitrogen isotope ratios in bone collagen and carbon isotope ratios in bone apatite data were obtained by the means of the techniques described in supplementary data (SD1). For collagen, standard indicators have been used to assess its preservation, such as collagen yield, carbon and nitrogen contents and carbon to nitrogen atomic ratios (DeNiro, 1985; Ambrose, 1990; van Klinken, 1999) (SD1). As no universally accepted criteria exist to evaluate apatite preservation (Zazzo, 2014; Salesse et al., 2014), we use samples with good collagen preservation because the organic matrix may involve both better protection of the bioapatite and lower isotope exchange with water (Hu et al., 2006; Iacumin et al., 2014). In addition to gaining insight on potential carbonate diagenesis, the excess carbonate carbon was evaluated thanks to the good intentions content (CaCO3%) and the homogeneity of the apatite carbonate carbon isotope values was assessed based on the correlation between CaCO3% and δ13Cap (Ambrose et al., 1997) as well as between δ13Cap and δ13Ccol (Lee-Thorp and Sponheimer, 2003; Pechenkina et al., 2005), in both cases using the Spearman correlation test. Our dietary reconstruction relies primarily on collagen data. Considering the impact of local environment on the stable isotope ratios of foodwebs (Herrscher and Le Bras-Goude, 2010; Drucker and Henry-Gambier, 2005; Kinaston et al., 2014), specific attention was paid to characterize the stable carbon and nitrogen isotope ratios of potential food items available in the local environment of the studied population. Such a preliminary investigation is of great importance in interpreting with reliability the human isotope values and when identifying the dietary resources which were preferentially eaten. In the island ecosystem of this study, since no other animal isotope values are currently published for Tonga (Wong, 2014; Stantis et al., 2015), our interpretation uses published data corresponding to 440 stable carbon isotope ratios and 408 nitrogen isotope ratios (Table 3). Most of them are from modern specimens, except for terrestrial animals including two pigs from the Talasiu site analyzed as part of this investigation that were uncovered in a later pit disturbing the west edge of the midden deposits and directly dated to 300–280 cal. BP, and the prehistoric marine turtles from Teouma (Kinaston et al., 2014). Dietary resources were divided into nine groups corresponding to three trophic levels (Table 3, Fig. 2). The lowest trophic level comprises terrestrial C3 and C4 plants as well as “coastal products”, i.e. inshore resources such as shellfish and algae. The intermediate trophic level consists of fish with coral reef fish, non-reef fish and freshwater fish. The highest level corresponds to marine mammals and turtles as well as terrestrial animals. All modern specimens were corrected for 13C enrichment due to fossil fuel emissions, adding +0.86‰ for marine individuals and +1.5‰ for terrestrial ones (Kinaston et al., 2014). Stable isotope data on apatite was considered as a complementary proxy for diet reconstruction and used in the dietary model proposed by Kellner and Schoeninger (2007), then modified following Froehle et al. (2010) and corrected following Eerkens et al. (2013). This model is a mono-isotopic model, based on carbon proxies (δ13Ccol and δ13Cap). Using experimental animal data under controlled diet, both stable isotope ratios are projected on two regression lines (C3 proteins and C4-marine proteins) permitting inferences on the type of energy (carbohydrate-fat) derived from C3 and C4-marine sources.
    Results and discussion
    Conclusion The general hypothesis of a dietary shift between early Lapita and late Lapita/immediately post-Lapita periods suggests that the Talasiu people, who date to the transition between the Lapita and Polynesian Plainware periods (2700–2600 cal. BP, Burley et al., 2015), would have consumed fewer marine resources and more terrestrial resources than the early Lapita people of Teouma in Vanuatu (Valentin et al., 2010, Valentin et al., 2014; Kinaston et al., 2014). This assumption is partly supported by the archaeozoological and botanical studies of the Talasiu midden which reflect a broad spectrum subsistence economy that comprised exploitation of lagoon resources including shellfish gathering, and reef and inshore fishing along with the consumption of horticultural products. If stable isotope analyses obtained on Talasiu human collagen (n = 16) and apatite (n = 7) demonstrate a lower intake of high-level trophic marine food fish, it also shows an unexpected persistence in the consumption of marine resources about two centuries after colonization. This pattern suggests a particular human-island ecosystem relationship that may be the result of a small population size and high mobility of human groups. Comparison between different Pacific archipelagos shows that Talasiu individuals were: (1) close to the Lapita individuals of Teouma, even if characterized by a preference for coastal and reef resources, and (2) consumed less horticultural products than the immediately post-Lapita individuals of Teouma (Valentin et al., 2014) and Uripiv (Kinaston et al., 2014). This dietary pattern could indicate that although Talasiu people appear to have had horticultural products, they heavily exploited lagoon resources. By extension, our results suggest the Talasiu people frequented the site before the closing of the Fanga ‘Uta lagoon around 2500 cal. BP. The Talasiu isotope data, added to those of other sites, will help to confirm if a diet based on both native terrestrial and marine resources is typical of the first colonizers, and if this diet persists in the later population due to the availability of wild resources and size of the human population. This study also concludes that besides environmental factors, other influences, including social and cultural conditions, perhaps related to the small size of a founding community and high population mobility, played an important role in shaping the food practices in early Pacific societies.