Across the vast Amazon basin, scattered patches of black and chocolate-brown earth stand in sharp contrast to the acidic, nutrient-poor red or yellow soils found in the rest of the rain forest. The dark soils are extraordinarily fertile, and the plant communities that grow there are different from those in the surrounding forest—higher in biomass, with a greater proportion of edible species like Brazil nuts and acai palms. Very often, they contain artifacts like ceramics or fragments of stone tools.
These soils are called terra preta, Amazonian dark earths (ADEs), or simply dark earths because they have been identified in Africa, Australia, Europe, and elsewhere outside Amazonia. They are also called anthrosols because almost all researchers agree that they were created by people. Dark earths have been found at hundreds of archaeological sites across the Amazon basin, covering an estimated 6,000-18,000 square kilometers. One study used modeling to predict that terra preta soils might actually cover more than 150,000 square kilometers, or 3.2% of the total forest.
Before Europeans arrived in the Americas—bringing conflict, exploitation, and infectious diseases that killed upward of 90% of the population—Amazonia teemed with human life. Archaeologists are still debating how many people lived there before the European conquest, but several estimates suggest they numbered between 6 million and 10 million.
Indigenous Amazonians made their homes on bluffs overlooking the rivers, where they fished and hunted and gathered and gardened. They domesticated plants, including manioc, sweet potato, and cacao. They carried out extensive earthworks, made roads, and modified wetlands. They set low-intensity fires to manage the landscape. And like humans everywhere, they produced garbage: fishbones, shells, manioc peelings, manure, weeds and crop residues, pottery, and charcoal.
Over generations, they turned that trash into treasure, creating rich, fertile earth good for growing crops. At the same time, the process sequestered large amounts of carbon in the soil. Scientists from a wide range of disciplines are now digging into terra preta for answers—not just for the new story it tells about Amazonia’s past but also for the lessons, possibilities, and warnings it might hold for Earth’s future.
Helping Plants “Grow Happy”
In the past, nonhuman explanations for the formation of terra preta have been suggested: sedimentation from floods, organic matter building up in lakes and small ponds, and ash fallout from Andean volcanoes.
In 2021, even, U.S. and Brazilian researchers published a paper proposing that the high fertility of Amazonian dark earths was the result of nutrients being deposited by rivers. Pre-Columbian peoples then identified these areas of increased fertility and settled there, the authors argue. “Indigenous peoples harnessed natural processes of landscape formation,” they write, “but were not responsible for their genesis.”
Most archaeologists, soil scientists, geographers, and anthropologists working in the Amazon, however, say there is little doubt that dark earths were made by humans. The prevailing winds blow the wrong way for volcanoes to be involved. Enriched soils are frequently found on top of bluffs—locations very unlikely to flood or collect water but wonderful places for people to live. Flooding can’t explain the wide variety of landscape types in which dark earths are found, why they are usually riddled with pottery fragments, or the fact that excavations most often reveal them within or around human-made mounds, pits, paths, and ditches.
The precise origins of terra preta remain unclear, but the lives of contemporary Indigenous Peoples in today’s Amazon give some insight into how these dark earths might have been made. Archaeologist Morgan Schmidt, a research affiliate at the Massachusetts Institute of Technology (MIT), has spent years researching the soils in partnership with the Kuikuro people of the upper Xingu River in Brazil’s state of Mato Grosso, whose ancestors have lived there for centuries.
To this day, Kuikuro farmers purposefully create dark earth for crop cultivation—they call it eegepe—and the composition of the soil and its spatial patterns are similar to those found around archaeological sites. The Kuikuro throw food and fire waste into trash middens around their houses, Schmidt said, and after a few years, plant crops and fruit trees on top.
“They’re constantly managing these plants and these crops in the backyard and improving the soil all the time,” Schmidt said. “They’re really proud of being able to make that fertile soil so that the plants can ‘grow happy,’ as they say.” Those modified soils end up rich in phosphorus, nitrogen, calcium, and carbon, with a pH much higher than other Amazonian soils.
One Indigenous colleague, Kanu Kuikuro, described the recipe to Schmidt: “Charcoal and ash we sweep, gather it up and then throw it where we will plant, to turn into beautiful eegepe. There we can plant sweet potatoes. When you plant where there is no eegepe, the soil is weak. That is why we throw the ash, manioc peelings, and manioc pulp.”
The Kuikuro also travel to nearby archaeological sites to grow crops in the older dark earths there. These days, they go by motorbike. Their contemporary middens now contain the odd battery and chunk of plastic. But in general, Schmidt thinks that the Kuikuro farmers are creating dark earth in much the same way their pre-Columbian ancestors did—the ancients just did it on a much bigger scale because there were so many more of them. But were they also making it deliberately, to improve the soil?
“There’s no way we can ever know what people were thinking in the past,” Schmidt said. “It’s very hard to find evidence of intentionality. But the Kuikuro people that we work with have demonstrated continuity in the area, and we find the same pattern of soil enrichment in the modern village and in the prehistoric sites, so we can be reasonably sure they were doing the same things in the past.”
Fertile Soils, Drought-Prone Forests
When the inhabitants of humid tropical areas are gone, most signs of human occupation are rendered invisible after 500 years. Wooden homes and ceremonial buildings rot away. Roads become overgrown. Mounds, ditches, and other earthworks are blurred by the vegetation, becoming visible only when the forest is clear-cut.
But below the surface, ceramic-strewn dark earth remains a tangible sign of human occupation. Finding out how much of it there is should shed light on the extent to which humans modified the rain forest, as well as how much carbon is locked away.
Mapping the extent of terra preta has been difficult, given the immense size of Amazonia, the remoteness of parts of it, and the thick forest canopy. (Many archaeological sites have, in fact, been revealed by Brazil’s accelerating deforestation.) Recent research, however, suggests that the shadow of dark earths can be seen from space.
Paleoecologist Crystal McMichael from the University of Amsterdam and her coauthors used satellite imagery to identify spectral signatures—differences in how light is reflected in forests growing on dark earth and those on other Amazon soils.
They proved that remote sensing can capture past forest disturbances and identify terra preta, but what they found surprised them. McMichael expected that fertile dark earths would support lush, green trees that were resistant to drought. The results showed the opposite: Forests growing on anthropogenic soil actually had less green canopy with lower water content compared with other soils, and these differences were accentuated in dry years, making the dark earth forests more fire prone and susceptible to drought.
There are several possible explanations for this discrepancy, McMichael said. Other studies have found that dark earths tend to support different kinds of trees, including a higher proportion of edible palm species, implying that pre-Columbian peoples changed the structure of the forest. Perhaps those areas of forest are still recovering from periodic controlled burns and clearing—500 years is only a couple of tree generations, after all.
Or it could be that these rich soils have continued to attract farmers for centuries, McMichael said. “The legacy of using them continues even today, and I think that’s part of why there are no giant lush things, and there are these shorter-stature palm-rich forests there.”
“Guardians of This Ancient Footprint”
It’s true that traditional farmers value dark earths, said ethnoecologist André Junqueira from Wageningen University in the Netherlands, but the relationship is not quite as simple as he expected when he began his research. Junqueira has studied how caboclos—present-day Amazonian peoples of mixed descent—cultivate the landscapes along Brazil’s Madeira River. The caboclos appreciate the high fertility of terra preta and plant some of their more nutrient-demanding crops and varieties there. But weeds, as well as cultivated crops, grow like wildfire in dark earth, Junqueira said, “so they actually require much more work to be maintained.”
“From a farmer’s perspective, what people like most is to have different types of soil that can sustain multiple cultivation systems and a wider portfolio of crops,” he said. The caboclo farmers were drawn to areas of the forest that have a mix of dark earth and ordinary soil—and the high concentration of useful trees and palms found near archaeological sites was a bonus, too.
By using these landscapes, they “maintained and amplified the pre-Columbian legacy,” Junqueira said. “In a way, they’re like guardians of this ancient footprint, and through their current practices, they’re still adding complexity and heterogeneity to the forest.”
But the caboclo farmers weren’t necessarily seeking out the most fertile soil, he suspected. In part, people have kept using these anthropogenic forests for the reason humans have always chosen their homes: location, location, location. “The same criteria that people use today to choose an area to live, they also used in the past: high bluffs, just on the margin of the river, close to a clean water source.”
That pattern is just what Schmidt and MIT geologist Samuel Goldberg and colleagues found in their own remote sensing and machine learning effort. By overlaying multiple bands of satellite imagery of the Xingu Indigenous territory, they could predict the areas of dark earths with reasonable accuracy. “We found a widespread pattern of ADE deposits located on river bluff edges on the uplands adjacent to the floodplains,” said Goldberg in a presentation at AGU’s Fall Meeting 2021.
In total, he said, the dark earths were predicted to cover 250–700 square kilometers, or up to 2.7% of the region. Multiplying this area by the carbon density and soil depths that have already been measured in field studies suggested that anthropogenic soils in the Xingu Indigenous territory could be sequestering 3–7 megatons of extra carbon in addition to the amount naturally stored in the soil.
Applying these densities to Amazonia as a whole—something Goldberg admitted is “simply speculation” at this stage—would mean that the dark earths of the Amazon could be locking up as much carbon as the amount emitted annually by the United States. Much more work needs to be done to find out whether that is really the case, said Goldberg, “but it does suggest that ADE could be a substantial reservoir of organic carbon in the soil.”
As carbon dioxide (CO2) accumulates dangerously in the atmosphere and nations struggle to feed growing populations sustainably, a technique that both sequesters carbon and improves soil for agriculture sounds like a silver bullet. And indeed, studies of terra preta have prompted a global research effort into the climate-mitigation potential of enriching soils with charcoal, creating a simplified modern analogue of dark earth, dubbed biochar.
Terra preta has “definitely been an inspiration” for biochar, said Johannes Lehmann, a biochar researcher and soil biogeochemist at Cornell University in Ithaca, N.Y., but the point is not to perfectly re-create it. The dark earths made by Indigenous Amazonians contain fish residues, manure, and ceramics, elements usually absent from biochar. Biochar has the focused goal of drawing down carbon dioxide from the atmosphere at the same time as improving agricultural yields.
The main technique to create biochar, called pyrolysis, involves charring waste vegetation at low temperatures in an environment with little to no oxygen. “If you starve it of oxygen, then you cannot oxidize a piece of wood to CO2 and water—and you leave a lot of carbon behind,” said Lehmann. The same piece of wood left to rot or burned in normal (oxygen-rich) conditions would release its carbon into the atmosphere within minutes to months. When vegetation is charred, the carbon remains trapped for decades or even centuries. “It is orders of magnitude more persistent.”
Recent research has found that biochar improves the soil in some contexts and has the most climate mitigation potential of any land-based effort, including agroforestry and afforestation—though Lehmann pointed out that nothing is more effective than keeping forests standing in the first place. More than 300 firms are already producing commercial biochar products.
Still, it’s early days for the industry. As a technology, biochar is about where photovoltaics were in the 1970s, Lehmann said. “In the ’70s, we all said photovoltaics will save the day, and then it took 40 years until anything happened.” Applying charcoal to soil might sound relatively simple—and indeed, it’s one of the few mitigation measures that can be started right now—but more research and practical experimentation are needed to determine where and how it can have the greatest effect, he said.
Most important, farmers themselves have to see a benefit in biochar, said Lehmann. “I don’t think we can expect an avocado grower to be a carbon farmer. If putting biochar on the avocado trees doesn’t improve the avocados, then frankly I don’t care whether it sequesters carbon. A land user will do it if she discovers that this is good for her soil.”
Scaling up will take time, then. “That’s the Achilles heel of a distributed system, right? It is more difficult to scale than a coal-fired power plant or an injection of CO2 into a big hole. But it also has beauty. It’s more robust once it’s there, and it’s likely more sustainable as it develops,” Lehmann said. “The worst thing that can happen is we have a few million happy farmers but have not saved the climate. I call it a no-regret strategy.”
Some researchers, however, have expressed concern that the commercialization of biochar could lead to illegal deforestation as biochar companies search for the cheapest raw materials to pyrolyze. In addition, biochar owes its existence to Indigenous Knowledges from Amazonian communities, they argue; if there are profits to be made, will Indigenous Peoples benefit?
Turning terra preta into a commodity could have implications for the Amazon’s ancient sites, too. Archaeologists remain deliberately vague about the precise locations of dark earth to protect those sites, said McMichael. In some parts of Brazil, terra preta sites are already being mined for potting soil or bulldozed to build modern towns. “There are hundreds of archaeological sites with dark earth being destroyed as we speak,” Schmidt said. “They’re protected by law, but there’s no enforcement.”
And they’re irreplaceable. Destroying these patches of precious earths entails a loss of history, of culture, of fertile cropland, and of biodiversity, Schmidt said. “Also, any carbon that’s stored in those soils will be emitted to the atmosphere.” The answer isn’t to lock them up, though. The best way to preserve them, he suspects, is for traditional communities to keep living on them, farming them with low-impact tools that don’t overly disturb the soil, and, through their actions, maintaining the millennia-old legacy of Amazonia’s human past.
Kate Evans (@kate_g_evans), Science Writer