In a finding that could reshape climate accountability, scientists have traced a significant share of the world’s most extreme heatwaves directly to the greenhouse gas emissions of individual fossil fuel and cement producers. The peer‑reviewed study, published in *Nature*, is the first to quantify the causal links between specific corporate emitters and the intensity and likelihood of heatwaves, applying advanced event attribution science to more than two decades of global data.
The study spanned 213 large heatwaves reported during 2000–2023 within the EM‑DAT international database of disasters those severe enough to result in widespread loss of life, economic loss, or emergency declaration. By applying the World Weather Attribution methodology, scientists blended observational data sets like ERA5 reanalysis with ensembles of CMIP6 Earth system models to contrast the climate under which those events took place with a hypothetical pre‑industrial counterfactual. The statistical framework was a non-stationary generalized extreme value (GEV) model, whose location parameter varied linearly with global mean surface temperature (GMST). This enabled the researchers to estimate the change in intensity as well as the probability ratio the factor by which climate change had amplified the probability of each event.
The outcomes were dramatic. Climate change intensified all 213 heatwaves and made them more probable, with median probability ratios increasing from around 20 in 2000–2009 to around 200 in 2010–2019. For 55 of the events, the rise in probability was more than a factor of 10,000, so they would have been extremely unlikely without anthropogenic warming. Median increases in intensity up to 2.2 °C were experienced in 2020–2023 compared to the 1850–1900 baseline.
In order to link these risks to emitters, the scientists used the OSCAR v3.3 reduced-complexity Earth system model in a residual attribution configuration. They deducted each of 180 “carbon majors” CO₂ and CH₄ emissions from past compatible emissions, and re-ran the climate simulations to calculate the change in GMST that can be attributed to each organization. These companies from investor-owned corporations such as ExxonMobil and Chevron to state-owned behemoths like Saudi Aramco and Gazprom are responsible for roughly 60 % of total anthropogenic CO₂ emissions since 1850, including downstream consumer combustion.
By combining the source attribution outcomes with the event attribution approach, the authors estimated the extent to which each firm had raised the likelihood and intensity of the heatwaves under investigation. The highest 14 carbon majors combined accounted for about half of the total climate‑forced rise in heatwave severity since the pre‑industrial era. Each company’s emissions were enough to increase the likelihood of at least 16 to 53 of the heatwaves by at least 10,000 times. The smallest among them was still responsible for more than a dozen such events.
Lead author Yann Quilcaille of ETH Zurich said, “We clearly quantify the causal link from their emissions to the intensity and the probability of occurrence of the heatwaves.” To illustrate, emissions attributed to Saudi Aramco and ExxonMobil each contributed 51 heatwaves at least 10,000 times more likely. At a mean level, the carbon majors’ emissions explained approximately 0.85 °F (0.47 °C) of the 2010–2019 median 3.02 °F (1.68 °C) intensity increase.
The research design fills a long-standing evidentiary gap in climate litigation. Legal expert and co-author Corina Heri explained, “Courts are indicating a willingness to hold carbon majors accountable, but at the same time asking for more scientific certainty, and our study helps to close a part of that gap.” By creating a quantified causal chain from identifiable emitters to identifiable adverse events, the research conforms to judicial requirements for a nexus of demonstrable fact between a defendant’s conduct and a plaintiff’s injury.
Attribution science has evolved in the last 10 years, from correlating global warming with broad trends to identifying human‑caused emissions’ influence on a single disaster. The probabilistic method employed in this study simulating thousands of runs of climate models with and without anthropogenic forcings has been developed to account for low‑probability, high‑impact events like the 2021 Pacific Northwest heat dome, which this research identified as 3.1 °C warmer and over 10,000 times more probable due to climate change.
Experts indicate that the implications go beyond the court. The results might guide policy talks, loss‑and‑damage financing schemes, and business climate risk assessments. As University of Oxford Sustainable Law Programme non‑affiliated expert Rupert Stuart‑Smith noted, “With more and more lawsuits aiming to hold high‑emitting companies responsible for their contributions to climate change impacts… work like this provides the basis for well‑informed judicial decision‑making.”
The authors do refer to limitations, such as heatwave underreporting in Africa and South America, and not treating other kinds of extremes like floods or wildfires, which they intend to look at in the future. But the technical innovation combining source attribution with event attribution on a global basis is a major step forward in being able to attribute climate responsibility scientifically.
