Temporary Overshoot of 1.5°C Could Trigger Amazon Dieback, Study Finds

New research appearing in Nature Climate Change suggests calamitous long-term danger of dieback for the Amazon rainforest if global warming surpasses 1.5°C, even briefly. The study evaluates the environmental consequence of passing the Paris Agreement's 1.5°C threshold, based on hundreds of climate-model simulations projecting outcomes for the Amazon and Siberian forests.

The study states even short overshoots would lead to mass death among the Amazon trees, slowly converting spots of the rainforest into dry savannah. The dieback process would upset ecosystems over several centuries and influence global carbon storage.

The research compared three paths of climate mitigation of the Sixth Assessment Cycle of the Intergovernmental Panel on Climate Change (IPCC). The "renewables" path eschews overshooting and keeps warming to less than 1.5°C by 2100. The "negative emissions" path allows overshooting through the application of massive-scale carbon dioxide removal technology to cool down temperatures. The "gradual strengthening" path features gradual policy advancement and brings global temperatures to 1.8°C at the turn of the century.

Scientists ran a basic climate model to calculate temperature results for each pathway, and they then simulated the response of forests. Land surface model JULES was then applied to examine forest health based on two metrics—net primary productivity (NPP) and forest cover. NPP is a measure of the rate at which vegetation deposits energy in biomass, a metric of forest health and growth, and forest cover estimates long-term structural change.

The outcomes show that under overshoot conditions, specifically in dry and hot environments, vast regions of the Amazon witnessed lower NPP and higher tree mortality. Between the negative emissions and renewables conditions, vast regions of Siberian and Amazonian forest witnessed lower productivity in the overshoot condition. In the Amazon, this means higher chances of dieback.

In all of the situations where 2100 global temperatures exceeded 1.5°C, 37% displayed signs of Amazon dieback. This rose to 55% in 2300 runs, showing that overshooting the threshold temperature rises long-term risk. Dieback conditions mostly arose in hot and dry climatic regimes, which are more likely when overshooting.

Conversely, the Siberian taiga reacted differently. Instead of undergoing high tree death, the area will likely face woody encroachment, or the transition from grassland to shrub and tree-dominance. The transition can impact carbon storage, water functioning and local diversity but is not as dramatic as Amazon dieback.

One of the most doubtful in the research is climate sensitivity—the degree to which the Earth temperature will react to increases in atmospheric CO2. Even under best-case mitigation conditions, if climate sensitivity was larger than taken for granted, even severe ecological hazards might still occur. The research points to the significance of preparing for a variety of climate futures, including low-probability but high-hazard ones.

Some limitations apply to the modelling used. Fire is excluded from the vegetation model and it fails to account for soil moisture stress, especially in tundra. These exemptions are likely to underestimate tipping point magnitudes, e.g., forest death or biome change irreversibility. In addition, while large-scale negative emissions are included in a few of the scenarios, their ability to deliver in the real world is not established. Relying on carbon removal technologies which are yet to be proven at scale introduces additional risk.

The results will play a critical role in the upcoming seventh IPCC evaluation cycle, including a new chapter on tipping points and other low-probability high-impact climate events. The research sheds light at a critical moment on the risk of even transient 1.5°C threshold crossings and the long-term implications for forests globally.

By examining an incredibly broad range of potential futures, the research finds that effort to prevent overshoot of 1.5°C must continue to be a top priority. The Amazon rain forest, which is both carbon sink and biodiversity hotspot, is especially responsive to warming, and its conservation is essential to global climate regulation. If it is allowed to warm more than 1.5°C—even briefly—the danger of irreversible change to the forests grows exponentially.

Source: Carbon Brief | Study published in Nature Climate Change, July 2025
Credit: Ayesha Tandon, Carbon Brief | Munday et al., 2025