Ancient food scraps provide clues to past rainfall in Australia’s Northern Territory


palm-like trees on a dry landscape

These pandanus trees, on Australia’s Magela Creek floodplain near the archaeological site of Madjedbebe, are the subject of new research involving the University of Washington.Florin et al., 2021, Nature Ecology and Evolution

Ancient food scraps found at Australia’s earliest site of human occupation, in the Kakadu region of the Northern Territory, are helping researchers generate rainfall records dating back 65,000 years.

A new study led by the University of Queensland and involving the University of Washington provides a glimpse into the region’s climate at the time when people first entered the Australian continent from the north.

Using the nutshell of the anyakngarra – also known as pandanus, a palm-like tree or shrub – the research team worked alongside Mirarr Traditional Aboriginal Owners to develop a novel method to investigate past rainfall at the archaeological site of Madjedbebe.

“Using the scraps from meals eaten tens of thousands of years ago, we can now tell a localised story of climate change and explore its effects on communities through time,” said the study’s lead author, Anna Florin, a research fellow at the ARC Centre of Excellence for Australian Biodiversity and Heritage and a doctoral candidate at the University of Queensland. “The nutshells hold evidence in their composition for the amount of water available to them when they were growing and can be used to understand past rainfall.”

The study was published Jan. 25 in Nature Ecology and Evolution.

The nutshells were discovered during excavations at Madjedbebe in Mirarr country in the Alligator Rivers region in 2012, and are the leftovers from meals eaten up to 65,000 years ago.

“The periods of low rainfall indicated by the pandanus nutshell correlate with increases in stone tool-making at the site, suggesting the region was a refuge during drier periods,” said study co-author Ben Marwick, associate professor of anthropology at the UW.

Data and code related to work at the site is openly available for other researchers, Marwick said, a critical aspect of archaeological study.

Coupled with other archaeological evidence from Madjedbebe, the research shows that the region was likely a good place to live even during glacial periods, allowing people to thrive during the driest spells in Australia’s history. This included the Last Glacial Maximum, between 18,000 and 25,000 years ago, an extended period of aridity around the world, Florin said.

The study also found that the modern era – not the Last Glacial Maximum – marks the driest time in recorded human occupation of the region.

“It is also a time when the plants and animals of the region are experiencing extreme hardships; with feral animals, loss of biodiversity and disruptions to cultural landscape management, including vegetation burning, all posing increased threats to the health and wellbeing of the landscape and its traditional owners,” Florin said.

Chris Clarkson, a professor at the University of Queensland and director of the excavation related to the study, called the research a huge leap forward.

“We’re now able to read the changing rainfall record through time and match this to the amazing strategies that were developed by Aboriginal people to cope with a dramatically changing landscape,” Clarkson said.

Gundjeihmi Aboriginal Corporation chief executive officer Justin O’Brien said the Madjedbebe site is providing an “extraordinary depth of knowledge.”

“This research reaffirms the importance of its long-term protection,” O’Brien said.

Other co-authors on the study were Patrick Roberts, Catherine Lovelock and Andrew Fairbairn of the University of Queensland; Nicholas Patton of the University of Canterbury; James Shulmeister of the ARC Centre of Excellence for Australian Biodiversity and Heritage; Linda Barry and Quan Hua of the Australian Nuclear Science and Technology Organisation; May Nango and Djaykuk Djandjomerr of the Gundjeihmi Aboriginal Corporation; Richard Fullagar of the University of Wollongong; and Lynley Wallis of Griffith University.

The study was funded by the Australian Institute of Nuclear Science and Engineering, the Australian Research Council, the Wenner Gren Foundation, the Dan David Foundation, the ARC Centre of Excellence for Australian Biodiversity and Heritage, and the Max Planck Institute for the Science of Human History.

/Public Release. View in full here.