The secret weapon that could finally force climate action

Arif Pujianto couldn’t sleep. All day, he had watched the waves rising from the nearby shoreline. When dawn broke, the water had risen further, and the area where he lives on the low-lying Pari Island in Indonesia was devastated: rubbish strewn everywhere, the walls of his home collapsing, the family’s drinking well polluted with salt water.

Since that night in December 2021, the tidal floods have returned dozens of times, making life almost impossible. “I feel angry and afraid,” says Pujianto. “If Pari Island sinks, where will we live?”

Eventually, he decided to do something about it. Pujianto is one of four residents of the island who have filed a lawsuit against cement manufacturer Holcim, demanding compensation for harms like these. At first blush, this might seem outlandish. After all, the company has no operations in Indonesia and is headquartered 12,000 kilometres from Pari, in Switzerland.

Yet Pujianto’s case is at the crest of a wave of litigation underpinned by innovative climate attribution models. Climate scientists say the most advanced type of model, called end-to-end attribution, can demonstrate a robust chain of cause and effect from an individual company’s carbon emissions all the way to local communities – no matter where they are.

Whether the studies will stand up in court is now being tested. “The science is evolving very rapidly and that’s allowing for new kinds of legal arguments,” says climate litigation expert Noah Walker-Crawford at the London School of Economics. What’s more, with the recent COP30 climate conference failing to deliver much meaningful action, some activists hope these advanced climate models could offer a powerful new weapon against global warming.

Computer simulations have been the backbone of climate science since the 1960s. Scientists run them to predict how the planet will warm as the amount of greenhouse gases in our atmosphere increases. This kind of research has informed, among other things, the reports of the Intergovernmental Panel on Climate Change (IPCC).

But climate models can be put to another use if they are run in a slightly different way. The idea is to simulate counterfactual scenarios and compare them with how things really turned out. How would the world look if we had left a certain portion of fossil fuels in the ground, for example? Climate scientists have spent decades using this technique to figure out the consequences of carbon emissions, in a field that is broadly called attribution science.

The first attribution simulations demonstrated beyond doubt the human origins of global warming. Then, starting in the early 2000s, researchers began to apply attribution science to specific extreme weather events. They would run simulations with and without human carbon dioxide emissions to see how the CO₂-elevated scenario affected the severity of particular heatwaves, floods and more.

The first of these studies, in 2004, revealed that the record-breaking heatwave in Europe the previous year was made twice as likely by anthropogenic emissions. Since then, similar links have been made to melting glaciers, wildfires, flood risk, ocean acidification and hurricane intensity. Today, the World Weather Attribution initiative routinely examines the extent to which our carbon emissions make specific extreme weather events more likely or more severe.

Over the past decade or so, researchers have been studying various parts of the causal chain in more detail, including the economic and health impacts on communities and the historic emissions of specific companies – but these different aspects of attribution science remained largely siloed.

In the past few years, however, all these ideas have advanced apace, and this made climate scientists Christopher Callahan at Indiana University and Justin Mankin at Dartmouth College in New Hampshire wonder if they could take attribution science further. They wanted to use the improvements to create an end-to-end climate model, one that stretches from the individual packets of emissions, perhaps from a single company or country, all the way to the effects of climate change on a community. “Those pieces all started maturing, so we could put them together,” says Callahan.

End-to-end attribution studies

Climate models can be extremely complex and often require supercomputers to run. The more granular you want to go, the sharper that problem becomes. But two advances have started to change that. One is the development of “reduced complexity” climate models, says Mankin. Instead of modelling Earth’s precise physical, chemical and biological processes – as the climate models that form the basis of IPCC reports do – these simpler models take a zoomed-out view and simulate them on average, which reduces processing power.

To ensure they remain accurate, reduced complexity models are continuously tuned by the results of more sophisticated models. The simpler models don’t require a supercomputer and can thus be run hundreds of times to explore a huge range of counterfactuals.

The second advance, according to climate campaigner Delta Merner at the Union of Concerned Scientists in the US, boils down to innovations in the way we account for emissions and join the links in the causal chain between emissions and harms. That includes, for example, smarter ways of differentiating between historical emissions. A gigatonne of carbon emitted today, when some carbon sinks are becoming saturated and oceans are more acidic, has different consequences to a gigatonne of carbon emitted in 1850. “If the cup’s full when you pour water in, that water is going to overflow,” says Mankin.

In 2022, Callahan and Mankin put these two advances together with others to produce the first end-to-end attribution study, which they used to draw a line between the emissions of various countries and economic damage caused around the world. They found that the top five emitting nations had collectively cost the global economy $6 trillion since the 1990s, with the brunt borne by low-income countries. “We showed for the first time that you could trace climate damages and warming back to individual country-level emitters,” says Mankin.

That was just the start. Perhaps the most comprehensive end-to-end attribution study so far, published in April 2025, looked at the emissions of specific companies – which includes direct emissions produced onsite and some indirect emissions, such as those produced when consumers burn fossil fuels. Callahan and Mankin built an end-to-end causal chain from the damages from specific extreme heat events back to the carbon emissions of individual firms. Applying their framework to all periods of extreme heat across the world between 1991 and 2020, they found the global economy had lost between $12 trillion and $49 trillion because of the emissions from the 111 most carbon-polluting companies.

They also ran their end-to-end climate model without the year-by-year emissions historically attributed to individual companies to estimate the impact each one had. This let them discern the difference a given firm’s emissions made. For instance, they calculated with 90 per cent probability that Chevron, the highest-emitting investor-owned company, caused between $791 billion and $3.6 trillion of economic losses over the same period. Chevron didn’t respond to a request for comment.

These studies focused on extreme heat and its consequences, as these are well understood. Mankin is aware that the research may be used in court and of the high bar that sets. “It better be bulletproof, right?” he says. The next step will be to start applying end-to-end models to other forms of extreme weather, such as flooding, wildfires and hurricanes, and to other economic and health impacts, which are a little more challenging to model because of the human factors involved and the lack of good data.

Climate litigation

Climate scientists usually want to remain strictly objective, to eschew any whiff of bias in their work. “Our objective is not to take down companies,” says climate scientist Yann Quilcaille at ETH Zurich in Switzerland. “We set out to fill in gaps in knowledge.”

Understandably, though, attribution science is attracting the attention of campaigners and activists. Any study that helps connect the dots between the emissions of certain firms and downstream harms may prove influential in court. The pinnacle of attribution science is end-to-end studies, which show how far the field has advanced and could prove even more powerful. “End-to-end studies are really the cutting edge,” says Walker-Crawford.

Climate litigation comes in many flavours and has been going on for years. So far, more than 3000 cases have been filed globally. One method is for someone to bring a case for damages against a company with massive emissions in the country where it is headquartered. These claimants are often people from low-income nations, who face some of the most severe consequences of climate change, while it is common for the biggest emitters to be powerful corporations in high-income nations. NGOs often support these cases. Litigators make pragmatic choices over which companies to target, as there are many “carbon majors” that could be deemed responsible. “They’ve been designed in a very strategic way, in the sense of seeing, ‘Where is there the highest likelihood of success?’,” says Walker-Crawford.

That is true for Pujianto, who is bringing his civil case against Holcim with his three co-claimants, supported by a Swiss aid organisation called HEKS. The plaintiffs and their lawyer decided to focus on Holcim because it is listed among the most carbon-polluting companies globally and because cement firms mostly produce “scope 1 emissions”, meaning the greenhouse gas is emitted while making cement. As such, there is no argument, as there would be for fossil fuels, over who actually produced the emissions.

A study commissioned for the lawsuit found that human CO₂ emissions were, with 99 per cent certainty, responsible for an additional sea level rise of between 16 and 26 centimetres on Pari Island during the December 2021 flood. In other words, water probably wouldn’t have entered Pujianto’s house without the emissions of carbon majors like Holcim. “Causality becomes much clearer and, by that, also the legal responsibility,” says Johannes Wendland, a legal advisor at HEKS.

For his part, Pujianto sees a clear logic. “The climate crisis is caused by the emissions of big companies, and Holcim is one of the biggest cement companies in the world,” he says. “People who emit less carbon must be protected and companies with huge emissions should be held accountable by law.”

In September, he travelled to Switzerland with his co-claimant Ibu Asmania. The Cantonal Court of Zug, one of the highest civil courts in Switzerland, heard about the climate harms on Pari and the claimants demanded that Holcim pay compensation and contribute to the cost of flood defences. They also said Holcim should cut its emissions in line with what is needed to keep global warming to no more than 1.5°C above pre-industrial levels.

In December, the court accepted the lawsuit, the first time that a climate lawsuit brought against a large corporation was admitted for trial in Switzerland. Holcim told New Scientist that the question of who is allowed to emit CO₂ and how much they should be allowed to emit should be a matter for legislatures, not civil courts, but added that it is deeply committed to taking action on climate change and has an evidence-backed plan for doing so that is aligned with a 1.5°C pathway. The company said it planned to appeal the court’s decision, but, other than the comment that it had already provided ahead of December, declined to add anything further.

There is a tangle of legal, moral and ethical questions that arise from all this, including who is truly responsible for carbon emissions – companies, households that use their products or the governments that are failing to effectively legislate against emissions? Merner notes that more than 70 per cent of historical global emissions can be attributed to just 78 companies (though some are or were state-owned). “You need to be focusing on actors whose actions materially influence global climate outcomes, and that is overwhelmingly the fossil fuel industry, not households,” she says.

There are tentative signs that the courts agree. In 2015, Peruvian farmer Saúl Luciano Lliuya brought a lawsuit against energy company RWE in a regional German court. Lliuya, a farmer and mountain guide, claimed that his home faced a risk of flooding due to a nearby glacier melting and demanded that RWE contribute to the cost of building flood defences.

Last year, Lliuya lost the case because the judges deemed that his house wasn’t at a high enough risk. Yet Walker-Crawford, who was a legal advisor on the case, says it was still a victory in a sense. “We didn’t end up having a full discussion about attribution in court, but the judges did recognise that there was very strong evidence on attribution.”

A proper test of how attribution science holds up in court in these “polluter-pays” cases hasn’t yet taken place, then. But such a test may not be too far off, says Walker-Crawford, especially given how many fresh cases are starting to come before judges. “There seem to be new cases filed almost every week at this point,” he says.

Yet his experience in court makes him well aware of the challenges remaining. One is that scientists and lawyers speak different languages. “Science and law are both trying to figure out the truth about the world in a broad sense, but they have different ways of doing that, so you need to find conceptual common ground,” he says. Often, successful lawsuits come down to telling a good story, which the facts of attribution science can form the narrative for, he says.

An even deeper challenge is the sweeping changes to the legal landscape that could turn the tide against these claims. In the US, for instance, President Donald Trump has been rolling back environmental regulations, which, to some extent, ties the hand of courts, says Mette Eilstrup-Sangiovanni, an expert in international relations at the University of Cambridge. “Litigation is only going to be as useful as the political and wider legal infrastructure allows.”

On the other hand, litigation may succeed where politics is failing, says Mankin. Pursuing climate action through the courts because international negotiations like COP are ailing is a tragic state of affairs, he says, but it may be necessary. He also wonders whether the rapid advance of attribution science could help tip the balance in climate lawsuits.

Indeed, if a precedent is set by even a single successful polluter-pays case, it could open the floodgates to a huge number of similar claims, says Walker-Crawford. In each case, the individual damages demanded may be relatively small – Pujianto’s damage claim comes to only $1334, which is less than Holcim’s CEO makes in an hour – but it could, nonetheless, be significant. “This is a small step,” says Pujianto, “but this small step can become much bigger.”