Skip to content

By Roger Highfield on

Amazon at a tipping point: the evidence

The Amazon, the world’s biggest tropical rainforest, shows signs of distress. Roger Highfield, Science Director, reports on new data that show how this remarkable ecosystem could turn into a savannah.

Hard evidence is published today that the Amazon is primed for a dramatic dieback which could release more planet-heating carbon dioxide into the atmosphere and disrupt the global climate system.

Scientists are worried that the Amazon rainforest could pass a tipping point in the near future. As in a game of Jenga, the dieback of the rainforest – a pivotal block of the global climate system – could trigger a wider collapse of the system and raise the risk of a planetary emergency.

In tropical South America both savannah and rainforest exist under similar climate conditions, the latter being a rich and self-sustaining ecosystem which is crucial for biodiversity, indigenous people, regional climate, and the global carbon cycle.

But when the great ‘flying rivers’ which transport large quantities of water from the Amazon Basin across South America dry up, a massive dieback of rainforest will turn this ‘crucible of biodiversity’ into savannah over a matter of decades.

The Amazon is also a key element in the global climate system because the vast rainforest has traditionally acted as a large carbon sink, though its ability to absorb the carbon dioxide linked with global heating has been in decline over the past 40 years.

However, during major droughts that killed many trees in 2005 and 2010, the Amazon temporarily turned into a carbon source. Recently, evidence has been published that the southeastern Amazon is now emitting carbon.

Today’s study in the journal Nature Climate focuses on a novel way to measure the ‘resilience’ of the Amazon and was conducted by Prof Niklas Boers of the Technical University of Munich and the Potsdam Institute for Climate Impact Research along with Dr Chris Boulton and Prof Tim Lenton of the Global Systems Institute, University of Exeter.

The team looked for empirical signs of ‘resilience loss’, evidence that a tipping point is approaching even though, at face value, the forest appears stable.

‘We looked for early warning signs,’ said Dr Boulton. ‘We found the amount of time the forest takes to get back to normality after a significant variation in weather, for example, is getting longer as the restoring feedbacks are degrading,’ he said, though he added the study is not able to predict when the tipping point will be reached.

The team quantified resilience by using high resolution satellite data to show the mass and greenness of vegetation, in other words its photosynthetic activity which reveals the extent to which it captures carbon.

Over two decades, they found significant declines in the south-eastern parts of the Amazon basin, along parts of the Amazon river, and in some northern parts of the basin.

Three quarters of the Amazon basin has been losing resilience since the early 2000s, notably in regions with less rainfall, and areas close to human activity. ‘We provide direct observational evidence that change, via increasing dry-season length and drought frequency, may already have pushed the Amazon close to a critical threshold of rainforest dieback,’ said Prof Boers. ‘This is the first time this has been shown by empirical evidence.’

‘We think resilience has been threatened by overall drier conditions which, unusually, culminated in three once in a century severe drought events (2005, 2010, 2015), combined with more land being taken over for farming, mining and so on,’ said Prof Lenton.

By the same token, the Exter/Potsdam study suggests that reducing deforestation will not just protect parts of the forest that are directly threatened, but also benefit Amazon rainforest resilience over much larger distances.

The study comes in the wake of the latest report from the United Nations’ Intergovernmental Panel on Climate Change (IPCC) which warns the threat that climate change poses is ‘unequivocal’. The report mentions tipping points and committed changes, when the rainforest may look healthy when it is actually committed to dieback. ‘We have seen changes in resilience, a link to this idea, which highlights the importance of looking at these indicators to assess how healthy the forest really is,’ said Dr Boulton.

Concerns about global tipping points dates back to paper written in 2008 by Prof Lenton, an advisor on Amazônia, an exhibition presented by Sebastião Salgado – currently transferring from the Science Museum to the Science and Industry Museum in Manchester – that showcases two hundred powerful black-and-white photographs that celebrate the indigenous peoples and varied landscapes of the Brazilian rainforest.

Archipel fluvial de Mariuá, Rio Negro, État d’Amazonas, Brésil, 2019 © Sebastião Salgado

Prof Lenton was inspired by the work of independent scientist James Lovelock. The Earth’s climate system is stabilised by negative feedback loops: as one example, higher temperatures stimulate more evaporation, and that means more clouds to reflect sunlight back into space, reducing temperatures.

However, there are positive feedback loops too, akin to the feedback or howl-round when a microphone is brought too close to a speaker. Now changes caused by warming, whether forest dieback or the thawing of tundra, will release more carbon and trigger more warming.

Two kinds of positive feedback are particularly important in the Amazon: forest degradation that staunches the flying rivers that transport moisture westward, reducing rainfall and forest viability. The second is localised fires, which amplify drought and associated forest loss by destroying trees so they may develop into ‘mega-fires’. Their new study suggests a weakening of the stabilising negative feedback mechanisms.

The team sought direct evidence because computer models do not agree on the timing and extent of these tipping point effects because they differ in terms of the detailed mathematical equations used to model the climate, and the numbers fed into these models. ‘Different models give different answers,’ said Prof Boers. ‘Though they give good answers for temperatures, they are less good at predicting the highly nonlinear dynamics relevant for the Amazon and its interactions with South American climate, including the flying rivers.’

Prof Boers is a coordinator of the EU-funded project TiPES (Tipping Points in the Earth System) that is looking also at other tipping points in the Earth system. Another study he published in Nature Climate Change last year reported that the Atlantic Meridional Overturning Circulation (AMOC), the Atlantic current system, may have been gradually losing its stability over the course of the last century.

The AMOC is responsible for the relatively mild temperatures in Europe but also influences weather systems worldwide. If the AMOC were to collapse, this would also have severe consequences on global and especially European weather and climate.

There would be increased cooling in the northern hemisphere, greater sea-level rise in the Atlantic, an overall fall in rainfall over Europe and North America, and further afield it could disrupt the tropical monsoon systems, including the South American monsoon.

However, global tipping points interact with each other in complicated ways. ‘Many of them have lost stability,’ said Prof Boers. ‘However, the slowing of the AMOC could increase rainfall in the dry season in tropical South America and might possibly increase the resilience of the Amazon. Sometimes, nature might be nice.’