Radiocarbon (14C) has been instrumental in clarifying how people came to inhabit the expanse of Pacific Oceania, now supporting an “incremental growth model” that shows a number of long-distance sea-crossing migrations over the last few millennia. A crucial step in this narrative involved the initial settlement of the remote-distance Oceanic region, in the case of the Mariana Islands around 1500 BC. The Marianas case can be demonstrated through delineation of stratigraphic layers, dating of individual points or features within those layers, redundant dating of samples in secure contexts, localized and taxon-specific corrections for marine samples, and cross-constraining dating of superimposed layer sequences. Based on the technical and methodological lessons from the Marianas example, the further steps of the incremental growth model will continue to be refined across Pacific Oceania. Many of these issues may be relevant for broader research of ancient settlement horizons in other regions.

Holocene climate change in the South Pacific is of major interest to archaeologists and Quaternary researchers. Regional surface ocean radiocarbon (14C) values are an established proxy for studying changing oceanographic and climatic conditions. Unfortunately, radiocarbon variability in the marine environment over the period of specific importance to human colonization of the remote Pacific islands—the last 3500 years—has been poorly studied. In order to build robust and accurate archaeological chronologies using shell, it is important to rectify this. In this paper, radiocarbon marine reservoir offsets (ΔR) are presented from eight archaeological sites, ranging in age from 350 cal BP to 3000 cal BP, and compared to coral datasets from the east Australian coastline. The results indicate that a significant decrease in the South Pacific Gyre ΔR occurred between 2600 and 2250 cal BP, most likely caused by changes in ocean circulation and climate. Accurately recording the timing of variability in reservoir offset is critical to untangling changes in society that took place in the Pacific, in particular, the development of Ancestral Polynesian Society.

We conducted a preliminary test of the coupled system of an elemental analyzer and the automated graphitization equipment Ionplus AGE3 (EA-AGE3 method) for accelerator mass spectrometry radiocarbon (AMS 14C) measurements of CaCO3 samples, by comparing with the conventional method where the samples are hydrolyzed in phosphoric acid and resulting CO2 gas is manually graphitized in a vacuum line (HPA method). The samples used in the test were the IAEA C2 travertine, fossil and modern corals from the Ryukyu Islands and the Ogasawara Islands, respectively (both are located in the northwestern subtropical Pacific). Results indicate that, relative to the HPA method, the EA-AGE3 method tends to cause an increase of ~0.4–0.5 pMC with more widely scattered data. This is presumably due to 14C contamination in the EA (the most likely cause seems to be a memory effect of 14C); this effect could be reduced by careful optimization of conditions and procedures in the EA process. The 14C data of pre-bomb annual bands (1931–1949 AD) in the modern Ogasawara coral obtained by the HPA method were used to estimate the marine reservoir 14C-age correction (ΔR) of this region; it ranges from –109 yr to –28 yr with the mean value with standard deviation of –81 ± 29 yr.

A Bison antiquus cranium and partial skeleton from Ayer Pond wetland on Orcas Island, San Juan Islands, Washington, date to 11,760 ± 70 14C yr BP. They lay in lacustrine sediments below peat, unconformably above emergent Everson Glaciomarine Drift (> 12,000 14C yr BP). Several bison finds in similar contexts on Orcas and Vancouver Islands dating between 11,750 and 10,800 14C yr BP indicate an early postglacial land mammal dispersal corridor with reduced water barriers between mainland and islands. New bison dates and published shell dates allow estimation of early postglacial relative sea-level trends for the San Juans, with a drop below modern datum ∼ 12,000 14C yr BP, and assist in evaluation of marine reservoir corrections. Emergence by ∼ 60 m is suggested by data from nearby areas. A tundra-like or meadow community and succeeding open pine parkland before 11,000 14C yr BP supported bison but horn-core reduction suggests suboptimal forage or restricted habitat. Expanding mixed-conifer forests after 11,000 14C yr BP contributed to bison extirpation. Dispersing ungulates such as bison must have influenced island vegetation establishment and early succession. Possible evidence for butchering by early coastal people adds significance to the Ayer Pond discovery, given its pre-Clovis age.

Radiocarbon dating of marine samples requires a local marine reservoir correction, or ΔR value, for accurate age calibrations. For the Samoan Archipelago in the central Pacific, ΔR values have been proposed previously, but, unlike some Polynesian archipelagoes, ΔR values seem not to vary spatially and temporally. Here, we demonstrate such variability by reporting a ΔR of –101±72 ΔR for the Manu‘a Group—the eastern-most islands in the archipelago—for the colonization period. This value is based on accelerator mass spectrometry (AMS) 14C and uranium-thorium (U-Th) series dating of individual coral branches from pre-2300 cal BP archaeological contexts. This figure differs from the previously proposed modern ΔR of 28±26 yr derived from dated historic, pre-1950, shell samples from the western islands of Samoa. Consequently, we recommend using the ΔR of –101±72 yr for the 1st millennium BC in Manu‘a, and 28±26 yr for calibrating dates within the 2nd millennium AD in the western islands (Savai‘i to Tutuila). Until more data from across the archipelago and from throughout the entire culture-historical sequence document ΔR variability, we recommend that researchers use both of these ΔR values to evaluate how the dates of marine-derived samples compare with AMS dates on identified, short-lived wood charcoal.