Conservation news

Ancient human sites may have distorted our understanding of the Amazon’s natural ecology

  • Scientists have traditionally based their knowledge of the Amazon rainforest on surveys from fewer than 1,000 plots of land, which they had assumed were representative of the rest of the forest.
  • Research now shows that many of these sites were occupied and modified by ancient peoples, and the trees are still regrowing from those disturbances.
  • These recovering trees absorb carbon at a faster rate than mature trees, so estimates of how well the rainforest can absorb carbon dioxide may be too high.

Pristine areas of Amazon rainforest are usually considered to be ancient, untamed jungles overflowing with old trees and biodiversity that have grown for centuries untouched by human hands. But that perception is starting to change.

Archaeological and agricultural evidence indicates this romantic idea may be a myth. An estimated eight to 20 million people once lived in the Amazon before their populations collapsed around A.D. 1500, when European settlers arrived. Now, a recent study suggests that human habitation left an imprint on the Amazon that modern ecologists have not fully taken into account when estimating the rainforest’s ability to recycle carbon or evaluating its biodiversity.

Many areas of Amazon rainforest are not as old or as undisturbed as was thought, the study shows. When today’s scientists examine the forest’s ecology, they are primarily looking within environments where ancient native peoples lived, cleared land, and cultivated crops. These relatively “young” areas of rainforest are still recovering from human occupation, so they are not representative of the entire Amazon forest.

Faster-growing trees in these areas may have led scientists to overestimate the amount of carbon the Amazon as a whole can store, the researchers state.

“Everything we know about Amazonian ecology and biodiversity comes from less than 0.0005% of the forest,” said Dr. Crystal McMichael, a paleoecologist at the University of Amsterdam in the Netherlands and the study’s lead author. The sites that scientists use to gain that knowledge are more likely to have been inhabited and altered by ancient people, she said. “That kind of skews our understanding of everything.” McMichael and team’s analysis appeared earlier this year in the Proceedings of the National Academy of Sciences.

The lush Amazon canopy appears wild and untouched by humans, but land was cleared and many tree species were cultivated by humans for thousands of years. Photo Credit: Rhett A. Butler.

McMichael and her colleagues compiled published maps of the plots used by Amazonian research teams to sample the rainforest. Modern scientists have used nearly 1,000 sites to gather data about trees and biodiversity. Of these sites, researchers have studied about 200 plots again and again to gauge how much carbon flows in and out of the forest’s vast array of trees. All told, these survey sites span about 250,000 hectares — a tiny swatch within the greater Amazon basin, which covers some 500 million hectares (or more than 1.2 billion acres).

The team then built a novel statistical model of where ancient Amazonian people probably lived. They considered areas of “terra pretas,” or dense rich soil good for farming; geoglyphs, or geometric earthworks; other architectural sites and lake sediments; and evidence of agriculture. Finally, they compared these predicted sites of human occupation, as well as previously known sites, to the data-gathering sites long used by ecologists.

The results showed that ecologists have disproportionately measured trees and other forest growth on sites that were likely to have been occupied by ancient peoples. Since humans have a huge impact on their environment, and rainforests take a long time to grow, this land is likely still recovering and does not represent the rainforest as a whole. Thus, the information we have about rainforest trees — which kinds are most common, how tall or densely they grow, how quickly they grow from season to season, and how much carbon they can store in their trunks, leaves, and roots — may not be accurate when ecologists extrapolate to the entire Amazon.

Brazil nut trees (Bertholletia excelsa), one of the “hyperdominant” species in the Amazon rainforest, were likely cultivated as food sources by ancient populations along with palms and fruit trees. Photo by Rhett A. Butler.

“This pioneering study of the rainforest challenges us to set a new benchmark for measuring things. It gives us a better chance to use the past to predict the future,” said Dr. Alexis Mychajliw, a conservation biologist at the Le Brea Tar Pits and Museum in Los Angeles, California, who was not involved in the study.

The clearest impact of the research is on our understanding of tree ecology in the Amazon. More than 200 species of trees are considered “hyperdominant” because they make up about half of the trees found in the rainforest. But many of these trees were probably cultivated on purpose, especially near where people lived, since they were useful as food or shelter. These people-associated trees are probably not as numerous in the rest of the rainforest, the study’s authors believe.

As a result, they maintain, scientists may need to start over to see the Amazon as it really is. Over the course of thousands of years, ancient peoples played a key role in the ecological development of the rainforest. Like other populations, they changed the land to suit their needs by burning, cutting, tilling, planting, and building. Before European-introduced diseases decimated the ancient native people around the year 1500, many parts of the Amazon were likely as cultivated as regions in Europe.

If some parts of the rainforest have been growing in their current wild state for only 500 years or so, it also changes what scientists can expect from the impact of this “new” growth on climate change. Ecologists believe the Amazon rainforest plays an important role in regulating the global carbon cycle by sequestering massive amounts of carbon released into the atmosphere by human activities. But since the Amazon testing sites used to measure this are probably still growing faster and thus sucking more carbon out of the atmosphere than the rest of the rainforest, scientists might be overestimating how much carbon the rainforest can store in the future, the authors state.

“When we think of the Americas, we tend to think of the baseline as after [Christopher] Columbus and the Europeans arrived,” Mychajliw said. But ancient people lived in the Amazon for thousands of years before that, dramatically affecting the landscape, she noted. “And, it is likely that we are underestimating how many people were actually living there back then.”

Paleoecologist Crystal McMichael, lead author of the study, conducting fieldwork in Ecuador. Photo courtesy of Crystal McMichael.


• McMichael, C. N., Matthews-Bird, F., Farfan-Rios, W., & Feeley, K. J. (2017). Ancient human disturbances may be skewing our understanding of Amazonian forests. Proceedings of the National Academy of Sciences, 114(3), 522-527. doi:10.1073/pnas.1614577114

Kimber L. Price is a graduate student in the Science Communication Program at the University of California, Santa Cruz. Other Mongabay stories produced by UCSC students can be found here.