Brazil Has a Second Amazon. We Almost Missed It.

The Amazon gets all the credit. It always has. When scientists talk about South America as a carbon sink, when policymakers argue about protecting tropical ecosystems, when satellite images of deforestation circulate on the news, it’s invariably the Amazon they mean: that vast, green, cathedral-dark tangle of trees holding, by some reckonings, roughly 150 billion tonnes of carbon in its wood and soils. The Amazon is the story. Or so everyone assumed.

Larissa Verona has spent several years pulling a different story out of the mud, sometimes quite literally. As a technician at Universidade Estadual de Campinas and a researcher now working with Amy Zanne at the Cary Institute of Ecosystem Studies, Verona has been drilling soil cores in Brazil’s Cerrado, the sprawling savanna that covers about a quarter of the country. She comes back from fieldwork caked in it, the Cerrado’s distinctive peat clinging to her boots and equipment. The work involves lugging a sensitive gas analyser called a LI-COR Trace Gas Analyzer through terrain that tends to swallow feet. “I fall over all the time in the field,” Verona says, “but as long as I protect the instrument, I don’t care. It’s like my baby.”

What she found in those cores, published today in New Phytologist, suggests the global carbon accounting has a rather large hole in it. Tucked inside the Cerrado, fed by groundwater rather than rainfall, are patches of waterlogged grassland that locals call campos úmidos and veredas. Wetlands, essentially. And their peaty soils, the team now reports, store roughly 1,200 metric tons of carbon per hectare. “This value is about six times greater than the average carbon density of biomass in the Amazon rainforest,” says Verona.

Six times. That number takes a moment to settle.

The mechanism isn’t complicated, though it took a while for anyone to think carefully about it in this context. Peat accumulates where water prevents oxygen from reaching dead organic matter, so decomposition stalls and carbon builds up layer by layer across thousands of years. Zanne puts it plainly: “Wet conditions create a lack of oxygen, which slows down decomposition. As a result, organic matter accumulates over time, enabling these wetlands to store large amounts of carbon in their soils, potentially for thousands of years at a time.” Colleagues at the Max Planck Institute for Biogeochemistry in Germany ran radiocarbon dating on the Cerrado peat and found the average carbon age was around 11,185 years, with some material that had been sitting there for more than 20,000 years. For comparison, the last great ice sheets were still retreating when some of this carbon first entered the ground.

That timescale matters enormously. Replanting a forest is, in principle, possible. Reaccumulating 20,000 years of peat is not, not on any schedule that means anything to the current climate crisis. “This carbon accumulated over a long time period,” Verona says, “and if it gets lost, we can’t accumulate it again for a very long time, the way we can replant a forest.”

Estimating the total scale of these wetlands proved more difficult than measuring what was inside them. The campos úmidos and veredas don’t form continuous swamps; they’re scattered in relatively small patches across the Cerrado, making regional estimates genuinely tricky. “Because these wetlands are naturally fragmented,” Zanne explains, “Larissa had to do lots of modeling to try to predict where they might occur.” The team combined remote sensing data with machine learning to map the likely distribution, and the result surprised them: the wetlands may cover about 16.7 million hectares, an area at least six times greater than previous estimates had suggested. That’s roughly 8% of the Cerrado, which in turn represents approximately 2% of Brazil. If the carbon density figures hold across that full area, the implications for global carbon budgets are considerable. The researchers reckon these wetlands could be holding the equivalent of about 20 percent of the carbon stored in all Amazon vegetation, though they’re careful to note that far more sampling across additional sites will be needed to test that figure.

None of this carbon appears in standard global carbon budgets, which is one of the more startling details of the study. “The Cerrado’s vast store of carbon isn’t included in our carbon budgets, because until recently, we didn’t know it was there,” says Zanne, adding that current protection strategies “tend to focus almost entirely on forests, and they overlook ecosystems like this.” There is a political dimension to this blindspot that Verona finds frustrating. The Cerrado, she notes, is widely regarded as Brazil’s “sacrifice biome”: the ecosystem that gets converted to soy fields and cattle pasture so that pressure stays off the Amazon. “Brazil wants to protect the Amazon,” she says, “but we also want to maintain agriculture. So agribusinesses are converting the Cerrado for commodity crop production, draining its wetlands, or taking its water for irrigation. And if we are removing the water from the wetlands, we expose these soils to more air causing greater breakdown and release of carbon.”

The problem is that the carbon, once exposed to air, doesn’t stay put. Verona spent time measuring greenhouse gas flux from the wetland soils across different seasons, pressing PVC collars into the ground and connecting them to her LI-COR analyser to capture what was coming up. Around 70% of the annual emissions occurred during the dry season, when the water table drops and microbes get access to carbon they’re normally locked out of. Under projected climate change, the Cerrado is expected to get hotter and drier; that shift, if it materialises, would push more of those ancient carbon stores into the atmosphere. There’s also a hydrological dimension that makes the sacrifice-biome framing particularly shortsighted. The Cerrado contains the headwaters of roughly two-thirds of Brazil’s major river systems, including, ultimately, the Amazon. “As we lose water in the Cerrado,” Zanne says, “we also put water supplies at risk across the country and broader region, including the Amazon. So, sacrificing the Cerrado for the Amazon actually puts the Amazon at risk, too.”

Brazilian law does technically protect groundwater-fed wetlands. But in some regions, as much as 50% of these ecosystems have already been degraded, the protections unevenly enforced and the wetlands themselves poorly mapped. The Cerrado’s problem, in part, has been invisibility: it doesn’t have the charismatic canopy of a rainforest, doesn’t photograph as dramatically, doesn’t register as obviously in the political imagination. It’s grassland, mostly. Easy to overlook.

Verona isn’t particularly interested in letting it stay overlooked. The team plans to continue refining their mapping and conducting more site-level carbon assessments to test whether the headline figures hold at scale. Other tropical peatlands, including systems in the Congo basin, store even more carbon per hectare (some exceeding 2,000 metric tons), so the Cerrado wetlands sit at the high end of the tropical range but aren’t quite record-breakers. What may be more significant is their vulnerability: spectroscopic analysis of the peat suggests the carbon there is chemically less stable than in wetter, more consistently saturated tropical peatlands, meaning it’s more likely to decompose quickly when conditions change. That combination of scale and fragility is what makes the new data urgent. “When we talk about carbon in Brazil,” says Verona, “we talk about forests. But the Cerrado is also important for its large and long-term carbon stocks, and we need to fight for it, too.”

What would it take for the Cerrado to get its moment in the global climate conversation? That probably depends, in part, on whether carbon accounting expands to include grassland peat, and whether international climate finance mechanisms start rewarding wetland protection the way they currently reward avoided deforestation. The science is getting sharper. The politics, for now, lags behind.

DOI / Source: https://doi.org/10.1111/nph.71027

Frequently Asked Questions