The Amazon tipping point—the critical threshold at which the forest no longer generates sufficient rainfall to sustain itself, leading to widespread dieback—is linked to key environmental changes including increases in deforestation, the advance of climate change, and the occurrence of forest fires. While both deforestation and climate change have been the focus of numerous research efforts and policy initiatives, forest fires have been much more neglected, especially by policy makers. This could be a result of forest fires being a relatively new issue in the region – historically, the Amazon did not burn. This Sprint will conduct research in this crucial area, and work alongside decision makers to develop policies and practices required to fireproof the Amazon.
This Sprint will bring together experts in ecology, remote sensing and social sciences, and work with representatives of the Brazilian Ministry of the Environment and local fire brigades and stakeholders.
Why this Sprint? Why now?
To fireproof the Amazon is an enormous environmental challenge, but, if achieved, it could avoid the tipping of one of Earth’s system, thus conserving its vast biodiversity and the livelihoods of millions of people. The burning seasons in 2023 and 2024 left decision makers at various levels of the Brazilian government (i.e. municipal, state, and federal) grappling for answers and calling for new research in this area. Additionally, the UNFCCC COP 30 will be held in Brazil this year, and results produced under this Sprint will be presented in side events at COP 30 in November 2025.
In 2026, the UK government will decide on its Seventh Carbon Budget, covering the years 2038 to 2042. Policies recommended by the Climate Change Committee (CCC) require widespread social change, additional investment of around 1% of GDP, and major changes in practices and technologies within many industries (CCC, 2025). To successfully navigate such major changes, the Department for Energy Security and Net Zero (DESNZ) have identified macroeconomic and distributional analysis as evidence gaps they must address to set the Seventh Carbon Budget. In this Sprint we aimed to develop new modelling approaches to fill those evidence gaps. The models assessed macroeconomic impacts of carbon budgets, including the impact on people with different levels of income and wealth. Methods were designed so as to be readily transferrable to other countries and regions.
Data-driven macroeconomic agent-based models have the characteristics required to model carbon budgets’ macroeconomic and inequality impacts. These models use real socioeconomic survey data to simulate the interactions of households, firms, government and the financial sector. This realism allows very specific policy interventions, in contrast to most macroeconomic climate modelling that flattens all policies into a carbon tax equivalent. This class of model is in its infancy but has already demonstrated the potential to rival the forecasting performance of DSGE* models (Poledna et al., 2023). Our team at INET Oxford developed a prototype model that covers all OECD countries (Wiese et al., 2024). As the model outputs individual income and wealth balance sheets for each of the thousands or millions of agents represented in each time period, it has great potential to conduct dynamic analyses of the distributional impacts of policy over time.
* Dynamic Stochastic General Equilibrium models
Why this Sprint? Why now?
Partnering directly with DESNZ, we aimed to provide provisional analysis in time to influence the 2026 Seventh Carbon Budget. This will enable macroeconomic and inequality goals to be taken into account when the government decides how emissions reductions are allocated between sectors of the economy, and how individual mitigation policies in the Carbon Budget Delivery Plan (DESNZ, 2023) are implemented.
