As global warming moves towards 1.5°C, understanding the risks associated with exceeding this threshold becomes more urgent. The impacts on human and natural systems are expected to increase with increasing increase with warming, and some may be irreversible. The latest report from the Intergovernmental Panel on Climate Change (IPCC) states that achieving 1.5°C of sustained global warming above pre-industrial levels is more likely than not by the early 2030s for all commonly used scenarios of future atmospheric greenhouse gas concentrations. However, future levels of global warming beyond this date are highly uncertain and strongly scenario-dependent.

Limiting global warming to 2°C or even 1.5°C above pre-industrial levels remains a societal aspiration and was proposed as a potential global policy at the COP21 conference in Paris. Depending on the strength of policy measures to reduce greenhouse gas emissions, warming could stabilize at or near 1.5°C, temporarily exceed it and return to that level, or exceed it and remain above it. Temporary temperature overshoots are considered potentially safe for some components of the Earth system, but even if they are compatible with long-term global warming goals, such an approach carries significant risks of spatially heterogeneous and potentially irreversible impacts. The IPCC report assessed the risks of overshooting in detail and showed that longer and higher degrees of overshooting increase the risk of potentially irreversible impacts, such as ecosystem and biodiversity loss.

The risk and magnitude of impacts of global average temperature exceedance and rebound on key forest ecosystems such as the Amazon and boreal forests are largely unquantified, particularly those from more realistic and policy-relevant emission scenarios. This study seeks to fill this gap using the PRIME modelling framework, which quantifies the consequences of policy-relevant exceedance scenarios for terrestrial ecosystems in a probabilistic and spatially resolved manner. Long-term impacts (to 2300) on two vulnerable ecosystems were analyzed: the Amazon rainforest and the Siberian boreal forests.

Three Illustrative Mitigation Pathways (IMPs) were selected that lead to long-term global temperatures close to 1.5°C, but with different trajectories before reaching stabilization.

• C1:IMP-Ren: A low-emissions scenario, achieving 1.5°C with little overshoot, primarily through extensive use of renewables.
• C2:IMP-Neg: A scenario with a larger temperature overshoot, relying on large-scale negative emissions technologies (CO2 removal) in the second half of the century, with a return to 1.5°C by 2100.
• C3:IMP-GS: A scenario with higher emissions and CO2 throughout the period, with warming around 1.8°C by 2100 and staying “well below” 2°C.

Key findings for the Amazon rainforest:

The Amazon rainforest is susceptible to a small but significant risk of long-term and irreversible die-off (dieback). Some simulations show a decline in primary production (NPP) of 10-20% or more, leading to a similar loss of forest cover. For regional warming levels above 2°C, forest loss is potentially significant. The findings suggest that Stabilizing global temperatures is not enough to stabilize the impacts of climate change on the AmazonOnly in the C2:IMP-Neg scenario, where the temperature drops significantly by 2300, does beginning of impact stabilizationWith decreasing temperature, forest cover stabilizes, which underlines long-term benefits of not ending at zero net emissions but continuing to remove CO2.

They were identified short-term and long-term "high-risk climate zones" for the Amazon, where there is a significant risk of die-off. The main factor driving these zones is regional temperature. At a regional temperature above 2.7°C in 2100, 58 % simulations recorded forest loss above current levels. This corresponds to a global temperature of 2.1 ± 0.5°C. In the long term (to 2300), 49 % simulations with a regional temperature above 1.7°C in 2100 (corresponding to a global temperature of 1.3 ± 0.3°C) recorded Amazonian die-off. Risk of Amazon extinction above 1.5°C global temperature increases with longer time horizon.

There is a low probability, but high impact, of significant Amazonian die-off even with the most aggressive mitigation policiesThe uncertainty associated with the sensitivity of the climate system significantly exceeds the differences between the selected IMPs scenarios, leading to a similar distribution of modeled forest losses by 2100 within each scenario.

Key findings for Siberian boreal forests:

The Siberian forest is probably destined for long-term and possibly substantial expansion of forest coverMost simulations assume a decline in NPP that is relatively modest even with a regional temperature increase of up to 7°C. All simulations show long-term increase in forest cover, which leads to irreversible changes in ecosystem composition (so-called woody encroachment). This afforestation has significant impacts on carbon sequestration, the hydrological cycle and biodiversity.

Scenario comparison:

The degree of climate change matters: the higher global temperature resulting from higher emissions in the C3:IMP-GS scenario leads to a greater risk of long-term loss compared to C1:IMP-Ren, which stabilizes just below 1.5°C. In the case of an exceedance (C2:IMP-Neg scenario), there is a clear the benefit of returning global temperatures to a lower level – even after exceeding the warming in C1:IMP-Ren, the C2:IMP-Neg trajectory shows a clear reduction in long-term binding changes in the forest.

Sources indicate that both ecosystems studied will almost certainly experience long-term changes, but especially The Amazon rainforest is at risk of significant lossThe risks associated with exceeding 1.5°C of global warming for the Amazon rainforest, although low probability, have high impact. These risks can be mitigated but not completely eliminated, and that by limiting the extent of warming and seeking to restore temperatures after any temporary excess. Immediate emission reductions and long-term investments in CO2 removal bring lasting benefits to forest health. Spring

The document was published in nature.com

Overview of key terms

• Global warming: Increasing the average temperature of the Earth.
• Pre-industrial conditions: A reference point (usually the second half of the 19th century) for comparing current temperatures.
• Paris Agreement: An international agreement that aims to limit global warming to well below 2°C above pre-industrial levels and to pursue efforts to limit it to 1.5°C.
• Overshoot: A situation where global temperature temporarily exceeds a target value (e.g. 1.5°C) and then returns to that level.
• Stabilization scenarios: Emissions scenarios in which global warming stabilizes at a certain level.
• Inevitable impacts: The impacts of climate change that can no longer be prevented.
• Return: The return of global temperature to the target level after a period of exceedance.
• Tipping point: A critical threshold beyond which large-scale and often irreversible changes occur in a system (e.g., an ecosystem).
• Hysteria: In the context of ecosystems, this means that the return to the original state after a threshold is crossed may not follow the same path as the degradation. In other words, the ecosystem takes longer to recover, or may not recover completely.
• Non-primary productivity (NPP): The rate at which plants in an ecosystem produce organic matter. It is an indicator of the health and productivity of a forest.
• Tree cover: Percentage of area covered by trees. This is a longer-term indicator of changes in the ecosystem.
• Woody encroachment: The expansion of woody plants into areas that were previously home to non-woody plants (e.g., tundra).
• Extinction (Dieback): Dying of trees or forest stands.
• Illustrative Mitigation Scenarios (IMPs): Socioeconomic scenarios that explore different paths to achieving the goals of the Paris Agreement. The article analyzes C1:IMP-Ren, C2:IMP-Neg and C3:IMP-GS.
• FaIR (Finite Amplitude Impulse Response model): A simplified climate model used to simulate profiles of CO2 concentrations and global average temperatures.
• PRIME (Probabilistic Regional Impacts from Model Patterns and Emissions): A modeling framework that links emission scenarios with regional impacts on ecosystems.
• JULES (Joint UK Land Environment Simulator): A land surface model that simulates impacts on terrestrial ecosystems.
• Scenario Models (ESMs): Complex models of the Earth system.
• CMIP6 (Coupled Model Intercomparison Project Phase 6): An international project that compares the outputs of different climate models.
• Climate sensitivity: The measure of how much global temperature will increase if the concentration of CO2 in the atmosphere doubles.
• Tail risks: Low probability but highly effective results