Refining dates for human habitation in the South Pacific

Associate Professor Fiona Petchey, Deputy Director of the University of Waikato Radiocarbon Dating Laboratory, is researching ways to refine the marine calibration curve used for converting carbon-14 values into calendar dates. Her team wants to help develop timelines that show how early peoples, including Māori, moved around and provide an understanding of natural and human impact on island ecosystems.

The problem and why we need a solution

Early in the use of radiocarbon dating, scientists began to see discrepancies between measured ages from the radiocarbon dating process (the conventional radiocarbon age or CRA) and known historical dates for artefacts. These discrepancies were in part caused by the fact that the atmospheric C-14/C-12 ratio, which is a key element in calculating radiocarbon ages, has not been constant throughout history. As a result, calibration curves were developed to enable experts to infer calendar dates from CRAs.

Currently, three curves are available and the limitations of applying these have become apparent over the decades. Fluctuations in atmospheric and marine carbon are often unique to different geographical locations.

Accurate dates are essential for scientists like those studying past climates. Some environmental scientists use shell and marine samples to look at past climates, environments and sea-level rise and fall.

Experts like archaeologists and anthropologists working to track the spread of people through the Pacific also need accurate dates to underpin their work. The study of human migration through Oceania and the South Pacific can provide insights into how early societies lived and adapted to climate changes, and it can enhance knowledge of wāhi tupuna (ancestral places).

Why shells?

Shells are a common material found in archaeological sites around the Pacific. Shellfish were a common food source – often discarded in large middens – but shells were also used to make decorative items and tools.

Shells present challenges for radiocarbon dating when interpreting a calendar age, so many researchers have avoided using this material. Fiona Petchey decided to take a closer look to see if there were ways to improve the calibration curve and offsets being applied to shells.

The challenges of dating shells

What makes shells so difficult to date using radiocarbon dating? Shells are found in freshwater, marine and estuarine environments. Within the estuarine environments, marine and terrestrial carbon mix, which means that neither a purely terrestrial (atmospheric) nor marine calibration curve will work.

To obtain a calendar age, it is necessary to understand how the ratio of C-14 and C-12 in the atmosphere and ocean varies geographically and over time. In addition, changes in ocean circulation and upwelling change the ‘age’ of the water in different regions. Ocean upwelling is caused when winds blowing across the ocean surface push water away from an area and cooler, ‘older’, more nutrient-rich water rises from beneath the surface where it may have been stored for thousands of years. Shellfish absorb some of this ancient carbon to make their shells.

Offsets in shell radiocarbon dates can also occur because of the ‘hard water effect’. This is when shellfish in limestone environments take up carbon from water that has drained through the limestone. Limestone forms from ancient calcium carbonate-based animals, and this ancient carbon can make shells living in these waters appear much older. Limestone landscapes and geology are common around Oceania, including in parts of New Zealand and Tongatapu as well as numerous smaller coral atolls.

Solving the shell issues

Fiona has approached her work from a number of different angles, including looking at combinations of different isotopes and dating shell growth rings.

To improve the chronologies of key settlement events, Fiona has selected samples from archaeological sites in the Mariana Islands and looked at different isotopes in the shells – not just carbon-14, but oxygen-18 and carbon-13. This approach was to see if she could detect a terrestrial signal in those shellfish that had lived in near-shore and estuarine environments. She hoped that, in an area where freshwater drained through limestone, a significant difference would be detected in the age of these estuarine shells compared with shellfish species that live further out in the ocean. Oxygen-18 can help to identify the temperature of the water, and carbon-13 identifies the source of the carbon input. By comparing all three isotopes, Fiona could separate shellfish influenced by cold marine water from those affected by warmer terrestrial water sources. She found errors of 300 years in the dates on some shells from the site studied.

As a scientist, Fiona had to look to see if she could reproduce the results at other sites. Looking at Tonga, she noticed a pattern. At specific times in the past, the magnitude of carbon-14 offset from the existing global marine calibration curve varied. Fiona plotted all the carbon-14 values and dates and identified that “something really strange or different” was happening between the ocean and the atmosphere between about 2,650 years ago and 2,350 years ago. This 300-year period is a significant time in the Pacific – it is when the Lapita people disappeared from the archaeological record, and soon after, a people termed ‘ancestral Polynesia society’ appeared.

Fiona then started to pull together radiocarbon dates from across the Pacific and could see patterns emerging at other times when significant cultural changes are apparent in the archaeological record.

Shell growth rings

Fiona is also investigating how shell growth rings might help to refine radiocarbon dates in the Pacific. Like trees, shells also put on growth rings that reflect the environment. As water conditions change due to tides and rainfall, the isotopes within these rings give us information about changing water conditions and potential sources of carbon that may influence the apparent age of the shell. However, the challenge of shell growth rings is that they’re very small.

What next?

Fiona and her team of archaeologists and anthropologists have received a Marsden Fund grant to continue this research. Fiona and her team have turned their attention to New Zealand, where there are thousands of radiocarbon dates from different sites. Looking at the terrestrial and marine dates from these sites, Fiona has also recognised significant differences between the atmospheric calibration curve and the marine signal shortly after ancestral Polynesian settlers arrived in New Zealand.

Results from some of this research are detailed in the article Ancestral Māori adapted quickly in the face of rapid climate change.

Fiona is excited by the possibility of improving radiocarbon dates during this period to advance our understanding of the Polynesian settlement of the Pacific.

The Aotearoa New Zealand Radiocarbon Database upgrade

The Aotearoa New Zealand Radiocarbon Database (ANZRD) is an open online resource containing conventional radiocarbon ages measured on archaeological materials around New Zealand. The database was set up over 20 years ago but had not been updated since then.

The need for consistent and accurate datasets for research and resource management has grown, so a team, including Fiona, worked to upgrade the site. This work included adding new ages, verifying previous datasets and drawing up new protocols to ensure a robust information source.

The upgraded ANZRD contains more than 4,100 C-14 ages from over 1,650 different archaeological sites and some museum samples.