It sounds like a heinous environmental crime. Deep in the Amazon jungle, Brazilian and British scientists are about to start releasing thousands of tonnes of carbon dioxide (CO2) into the atmosphere from almost a hundred giant towers erected amid the trees.
But dastardly as it sounds, the $50m (£36.5m) project’s purpose is benign. Backed by their respective governments, the researchers are embarking on a decade-long experiment to create future atmospheric CO2 levels.
Their aim? To finally answer one of the most critical questions for the future of the Amazon rainforest and Earth’s climate.
They want to know if most current climate models are right to predict that the extra CO2 in a future atmosphere will fertilise tree growth and so soak up some of our emissions – buying us time to stop burning fossil fuels – or whether that hope is misguided.
Perhaps instead, the forest will die off, releasing CO2 and supercharging global warming later this century.
The Amazon Free-Air CO2 Enrichment project (FACE, for short) will release the greenhouse gas into the Cuieiras Biological Reserve, some 70km (almost 45 miles) north of the rainforest city of Manaus. It began in May 2026 and will keep going day-in and day-out until 2036.
A joint project of Brazil’s National Institute for Amazon Research (INPA) and the UK Met Office, Amazon FACE is “one of the most exciting experiments on the planet,” says Prof Richard Betts, head of climate impact research at the Met Office.
“The Amazon is a key uncertainty in calculating global carbon budgets,” says his Met Office colleague and co-British science lead on the project, Dr Andy Wiltshire. The project “will provide vital real-world data to reduce these uncertainties.”
Their 96 tree-high towers are primed to begin spraying CO2 into the air around some 1,300 rainforest trees of more than 400 species, hoping thereby to stimulate accelerated photosynthesis, the chemical process in leaves that uses sunlight to convert CO2 into plant matter.
The aim is to raise CO2 levels during daylight hours to around 630 parts per million (ppm). That’s 200 ppm above current levels – the potential global norm by around 2080. In effect, the experiment is a time machine for the atmosphere and a stress test for the world’s largest rainforest.
Disagreement over CO2
The future of the Amazon and other tropical rainforests is one of just a handful of topics where climate modellers are sharply divided.
Most of their models predict that the forests have a bright future, with rising levels of atmospheric CO2 boosting photosynthesis so the trees will grow faster. They call it CO2 fertilisation. By absorbing CO2 from the air, future trees could both grow the rainforests and slow global warming.
However, a growing number of scientists are sceptical. They predict that any fertilisation boost will be negated by a shortage of nutrients in Amazon soils.
Faced also with rising temperatures and worsening droughts, trees could instead cross tipping points beyond which they die off or give way to arid savannah. That would create a dangerous climate feedback, with rising temperatures releasing more CO2 from the forests, which would in turn raise temperatures further.
The Amazon FACE project’s primary job is to discover which forecast is right, says Betts. “Will photosynthesis increase as much as predicted in models… or will the responses be different [in] a forest exposed to the elements and other [unpredictable] environmental factors?”
To help assess those wider factors, the project will look at a range of other potential effects of the extra atmospheric CO2, he says. For instance, it’ll examine how the CO2 boost will influence the way trees use moisture.
It could allow trees to use water more efficiently and release less into the atmosphere, a process called transpiration. While this might help individual trees withstand periods of drought, it could also dry out the air above, ultimately reducing rainfall across the Amazon.
The scientific jury is still out on such questions. “No single experiment can expose a system to all the changes that may occur in the future,” admits Prof Iain Hartley, an ecologist of the University of Exeter and lead author of the current science plan for the project.
So, talk of the FACE project being a time machine for the Amazon may be an oversimplification. But he says the experiment could dramatically improve existing models and help resolve whether the forces of fertilisation or savannisation will win out.
The drama about to unfold
The scene is now set for the scientific drama ahead. The CO2-spraying towers rise up like bizarre metallic trees amid the natural forest.
The 35m-high (114ft) towers have been erected in six rings, each 30m (almost 100ft) in diameter and surrounding dozens of mature trees.
The towers comprising three of the rings will pour CO2 into the air, while those in the other three will act as controls, releasing ambient air. Each ring has a crane reaching above the towers to carry out maintenance and monitor the canopy from above.
The CO2 will be delivered in liquid form to 25-tonne pressurised tanks. Operators will vaporise the liquid as required and send it down pipes to the towers. Each tower then has two vertical pipes stretching from the ground to the forest canopy that will release the gas through perforations.
The experiment will be computer-controlled to ensure CO2 levels are consistent within the rings. Real-time monitors will constantly check CO2 concentrations, along with the speed and direction of forest winds that will inevitably disperse the gas into the wider atmosphere.
Using this data, algorithms will calculate how much CO2 needs to be added, and control the releases by altering the speed of flow of CO2 up the pipes, opening and closing the perforations as necessary.
Release rates will vary, but should average between 2.5 and 3 tonnes per day, says Dr David Lapola, an ecologist at Brazil’s State University of Campinas, whose scientists are running the project along with INPA and the Met Office.
To find out how the forest behaves in the CO2 time machine, his researchers will measure a range of key indicators of forest health. This will include the rate of photosynthesis and storage of carbon in the wood, roots and leaves of the trees, as well as the cycling of carbon, nutrients and moisture in the soil.
They will also look for any emerging changes in the species makeup of the Amazon. Which species will be the winners and which the losers in the time-warped greenhouse rainforest?
Read more:
- A global food crisis is nearing – and the warning signs are buried beneath your feet
- Deadly fungal storms are now sweeping the US – and spreading a disease few doctors recognise
- The most powerful climate phenomenon on record could hit the US in 2026, experts warn
The struggle for answers
It’s frustrating for scientists that they still don’t have clear answers to such basic questions. But they’ve long been thwarted by the failure of governments and research agencies to invest in the necessary large-scale analysis.
The first calls to carry out real-world experiments with elevated levels of CO2 in a tropical rainforest were made as long ago as 1991 by Prof Harold Mooney, an ecologist at Stanford University and advisor to successive US governments.
But researchers were constantly dogged by a lack of funding and the necessary government buy-in, until 2014, when the Amazon FACE project finally got financial backing from the Inter-American Development Bank (IDB).
Within weeks, the scientists were busy, choosing a site for the research and starting the collection of baseline data on the current state of the designated patch of forest.
Later, the IDB got cold feet, but the British government stepped in, as part of its commitment to climate science ahead of hosting the 2021 UN climate negotiations in Glasgow. Its £7.3 million (almost $10m) paid for the now-completed construction of the towers and other infrastructure.
But even to its last stages, the task has been hampered by logistical problems. “Ironically, climate change itself made the construction process take longer,” says Wiltshire.
“Major droughts two years in a row [2023 and 2024] stopped us getting equipment shipped in because river levels on the Amazon were too low.”
During the decades-long gestation of the Amazon FACE project, researchers set up other open-air CO2-enrichment projects outside the tropics (see ‘A breathalyser for the rainforest’, below).
The first experiments tested young trees. They mostly showed a positive response, with saplings growing ever faster the more CO2 was added.
The next question to be answered: whether mature trees would do the same. Many experts thought they wouldn’t, says ecologist Prof Richard Norby of the US Oak Ridge National Laboratory, the mastermind behind a series of experiments aimed at finding out.
He fumigated mature plantations of sweetgum trees in his lab’s research park among the temperate forests of Tennessee and then helped establish a British study. This has been carried out by the University of Birmingham’s Institute of Forest Research (BIFoR) in a 180-year-old oak woodland in Staffordshire and is set to conclude later in 2026.
In both cases, Norby and his collaborators were surprised to find that mature trees responded in the same way as young trees, with a 40 per cent CO2 enrichment bringing an average increase in photosynthesis of just over 20-per-cent.
“The response declined with increasing mean temperature, but didn’t decline with forest age,” Norby concluded in a 2025 review.
But another FACE experiment in Australia had different results. Adding CO2 to a hundred-year-old eucalyptus forest in the Blue Mountains near Sydney produced no extra growth.
Prof David Ellsworth of Western Sydney University, who has been running the experiment, says the problem is that the local soils are short of phosphorus, a key nutrient necessary for photosynthesis. This shortage has thwarted any CO2 fertilisation.
Instead, the extra CO2 ended up fertilising the growth of soil microbes. “This left less phosphorus available for the trees to take up,” he says.
That finding raised alarm bells, since most tropical rainforest soils, including those in the Amazon, are also low in phosphorus. So, would they too succumb to a phosphorus famine and fail to grow faster in the future CO2-rich air?
Ellsworth warns against reaching simple conclusions. His experiment involved just one type of tree: eucalyptus.
“The high diversity of Amazonian trees means that some should be able to gain phosphorus efficiently and so grow better with increased CO2 , but a majority will probably behave like our eucalypts and not increase their carbon storage,” he predicts.
“Our global estimates for the future for carbon storage in these forests could be too high,” but he hopes the Amazon FACE project can provide the answer the world badly needs.
All set... but for the CO2
Back in the Amazon, Lapola still faces logistical challenges. He calculates that his experiment will need some 3,300 tonnes of gas to be expelled into the forest air each year from the towers, making some 33,000 tonnes over the coming decade.
But finding a secure supply of such large volumes of CO2 to be delivered deep into the rainforest is proving difficult and expensive.
Just weeks away from the project’s operational start, his team was unable to say for sure where the CO2 will come from. But the plan is to have several different sources to improve reliability. They’ll probably include established producers in Manaus that supply the fizz in the city’s Coca-Cola.
Money is also an urgent issue. At a likely cost of $1,000 (around £730) for every tonne of CO2 , the bill for the gas needed over the next 10 years will probably be more than $30 million (almost £22m). Seeking funding is “an ongoing process,” says Betts.
It’s far from clear if the British government, which has recently reduced aid funding, including for environmental activities, will put up more cash. If it doesn’t, its megaproject could languish unused deep in the rainforest. “I’m not aware of a Plan B,” says Norby. “Except, I guess, to be optimistic.”
But if all goes well and the gas begins pumping in May, he expects reliable results from the experiment within two years.
But what about that extra CO2 being pumped into the atmosphere? Well, a few tens of thousands of tonnes emitted in the name of crucial climate science hardly stands comparison with the tens of billions of tonnes put up each year from burning fossil fuels.
Amazon FACE’s contribution will be equal to a bit more than the annual emissions of a British town of around 7,000 people. Or a cola bottling plant.
A breathalyser for the rainforest
While the towers of the Amazon FACE project are set to probe the effects that altering the air has on the trees, a far taller tower nearby is already sniffing the air above the rainforest to find out the opposite: how the trees are affecting the atmosphere.
The Amazon Tall Tower Observatory is the highest manmade structure in Latin America. It stands 325m high (over 1,000ft), roughly as tall as the Eiffel Tower, on a remote riverbank about 100km (62 miles) east of the Amazon FACE site.
It has been analysing the air above the forest since 2015. I went to see it in 2019.
The researchers in charge of the tower had attached sensors all the way up to sniff the forest’s breath. They were most interested in trace gases called volatile organic compounds (VOCs), such as isoprenes and terpenes, being released from pores in the leaves of the trees underneath.
These VOCs act as condensation nuclei for water droplets in the air above. These droplets coalesce to form clouds, recycling the moisture released by transpiring trees.
In this way, the VOCs help create rainfall downwind that sustains rainforest trees for hundreds of kilometres, all the way to the Andes. The chemistry creates a symbiotic relationship between the trees and the climate: just as the rainfall is necessary for the rainforest, so the rainforest is necessary for the rain.
At the top of the tower, I met the mastermind of the project, Meinrat Andreae, director of the Max Planck Institute for Chemistry. “There are reactions going on in the air that we haven’t accounted for yet,” he told me.
And he had a hunch that the key to the mystery was a subset of little-known VOCs called sesquiterpenes.
These are extremely reactive, disappearing into new compounds within seconds of their release from the leaves. He believes that their reactivity made them potentially big-time cloud creators.
Six years on, it turns out he was right. Andreae’s researchers have since shown that Amazon trees are a major source of these elusive sesquiterpenes, and they’re vital for recycling moisture in the Amazon.
Intriguingly, they’ve found that the trees emit more sesquiterpenes during hot, dry weather. They seem to be designed as potent drought-breakers – a way for the trees to create rain and help the forest survive.
That could make them critical for the Amazon rainforest’s fate in a greenhouse world. And another factor to weigh up in the great debate about the future of our climate.
Read more:
