By Early settlers of the island of Rapa Nui are famous for having created massive stone statues. They have also gotten a bad rap as instigators of a population boom that led to ecological and social disaster.
A new analysis of the island’s landscape suggests Polynesian seafarers who reached Rapa Nui around 800 years ago maintained a modest farming system and a small but stable population of no more than around 3,900 individuals until Europeans showed up in 1722 and renamed the early settlers’ homeland Easter Island, the scientists report June 21 in Science Advances.
That finding challenges a popular idea that expanding Rapa Nui communities grew so large that they exhausted available resources, decimating the island society by the time Europeans arrived in what’s been called an “ecocide” event, say archaeologist Dylan Davis of Columbia University and colleagues. The new study’s population estimate is in line with accounts written by early European visitors to the island that suggested that about 3,000 people lived there.
But some Rapa Nui researchers contend that that the new investigation relies on data that are too limited to draw any conclusions about how much food was grown on the island and how many people agricultural activities could have supported.
It has been difficult to estimate how many early settlers inhabited Rapa Nui because knowledge remains limited about different farming practices on the island and the range of crops that were cultivated.
Two prior studies by other researchers used various climate and soil measures to address that agricultural mystery. Researchers know crops were grown in rock gardens, where the rocks helped to enrich volcanic soil(SN: 12/16/13). Strategically placed rocks and boulders protected cultivated sweet potatoes from wind, prevented evaporation of rainwater, minimized weed growth and increased soil nutrients. But researchers have yet to determine the total number of rock gardens and all of their locations.
One investigation, which included satellite-gathered data generated from near-infrared light wavelengths that distinguish between rocks and vegetation, estimated that from 2.5 percent to 12.7 percent of Rapa Nui’s roughly 164-square-kilometer surface was covered by rock gardens. A second study suggested that about 19 percent of the island was suitable for growing sweet potatoes and populations could have ranged from around 3,000 to 17,500 or more, depending on crop yields and other factors.
In the new study, Davis and colleagues first conducted ground surveys on Rapa Nui in 2019 and 2023 to identify chemical and geological features that characterized rock gardens, bare soil, grassy areas, forests and other parts of the landscape.
Using that data, the team trained three machine-learning models to use that ground survey data to identify rock gardens across the island in a newly available set of satellite images detailed enough to identify differences in vegetation and soil composition. Those images portray different shortwave infrared wavelengths that can detect signature features of rock gardens, such as spots on the landscape with higher moisture and increased nitrogen levels. The best-performing machine model correctly identified rock gardens that had already been recorded in ground surveys around 83 percent of the time.
The analysis revealed that rock gardens covered less than one-half of 1 percent of Rapa Nui’s territory. That estimate amounted to less than 0.76 square kilometers of rock gardens on the 164-square-kilometer island.
A rough calculation of the nutritional value of sweet potatoes grown in rock gardens, supplemented with fish, other foods from the sea and less nutritious crops such as bananas, indicates a maximum Rapa Nui population of 3,901 individuals, give or take about 800, Davis estimates.
“Rock gardening made Rapa Nui soils productive enough to grow sweet potatoes, but we don’t find that this cultivation technique was sufficient to support large numbers of people,” adds archaeologist Carl Lipo of Binghamton University in New York.
But the new study substantially underestimates how much of Rapa Nui was covered by rock gardens, and thus how many people could have lived there in pre-contact times, says archaeologist Christopher Stevenson of Virginia Commonwealth University in Richmond. Stevenson coauthored the two earlier studies of Rapa Nui’s rock gardens and population sizes and did not participate in the new study.
Davis and colleagues trained machine models by using environmental data collected near Rapa Nui’s coastline, says Stevenson. Training data did not include evidence from surveyed rock gardens in upland parts of the island, he contends. Rock gardens dot the mid-slopes of Rapa Nui’s largest volcano, “but according to the new paper, there are no rock gardens there,” Stevenson says.
A comprehensive evaluation of past population sizes on Rapa Nui has yet to be conducted, Davis says. Even if rock gardens occurred in larger numbers than assumed in the new report, his team considers their pre-contact population estimate a solid guide for further investigations.
The new paper also does not account for Rapa Nui cultivation practices other than rock farming that boosted agricultural productivity, say archaeologists Sarah Sherwood of the University of the South in Sewanee, Tenn., and Jo Anne Van Tilburg of UCLA in an email to Science News.
For instance, Sherwood and Van Tilburg’s excavations at an inland site where statue carving occurred have revealed signs of plant cultivation in exceptionally fertile soil that did not require rock gardens (SN: 10/25/19).
Another problem concerns the inability to determine whether rock gardens considered in the new study were used simultaneously or at different times, Sherwood and Van Tilburg say. Davis and colleagues cannot exclude the possibility that “rock gardens were sparsely used or unsuccessful adaptations that inadequately fed a fast-growing population.”
Everyone agrees that advanced satellite imagery informed by continued investigations of cultivation areas on Rapa Nui will bring pre-contact human numbers into better focus.
