The true costs of climate disasters
Scientists have linked extreme weather to climate change. Now they can quantify its impact on the damage, writes Chloé Farand.
It took just two days in September 2024 for Hurricane Helene to intensify from a tropical storm into a major category 4 hurricane, the strongest to make landfall in Florida’s Big Bend region since reliable records began.
Helene brought violent winds, record-breaking storm surges and deadly flooding to the southeastern United States. It killed at least 250 people – the deadliest hurricane to strike the US mainland since Hurricane Katrina in 2005.
Homes were flattened, roads flooded, cars washed away and millions were left without power.
A fortnight later, the western coast of Florida was battered by another tropical cyclone. Milton made landfall as a category 3 hurricane, spawning dozens of tornadoes in its wake. More than 40 people were killed in the US and Mexico.
In the couple of days that followed, researchers at Imperial College London estimated that, in Florida, 44% of the direct economic damages caused by Helene and 45% of those caused by Milton can be attributed to human-induced climate change.
Using pioneering methodologies, the studies were the first to pinpoint how much of the economic damage of a cyclone was caused by human activities so quickly after it made landfall.
This novel and fast-developing field of attribution science is known as impact attribution. It can help communicate the cost of climate inaction in a way that is relatable to people’s experiences of a warming world, explains Dr Emily Theokritoff, a research associate at Imperial’s Grantham Institute – Climate Change and the Environment, who contributed to the studies.
“The main motivation is to raise awareness of the cost of climate change and to illustrate how much worse the impacts are,” she says.
In recent years, scientists at Imperial have led efforts to develop global methodologies to identify the influence of climate change on extreme weather events.
The World Weather Attribution (WWA) initiative has carried out hundreds of rapid attribution studies of disasters that hit all corners of the planet. This work is helping answer questions such as whether heatwaves, droughts, floods and wildfires were made more likely, more intense, or lasted longer because of climate change.
But a multi-disciplinary team of Imperial researchers wanted to go a step further in the causality chain and determine the extent to which climate change is responsible for the damages caused by any given extreme event.
They argue that such attributions could help society better prepare for climate shocks, inform discussions on responding to the losses and damages endured by the most vulnerable communities and even help bring the world’s biggest polluters to face justice for the impacts of their activities.
Inspired by the model established by WWA, Dr Theokritoff is leading the development of a global and rapid impact attribution framework using peer-reviewed methodologies to estimate the additional economic and human impacts that storms, heatwaves and floodings bring due to climate change.
The methodology to attribute the economic damages of tropical cyclones is the first to be operational.
Tropical cyclones, which are also known as hurricanes or typhoons depending on their location, are among the deadliest and most damaging extreme weather events.
Climate change is making the most powerful of these large storm systems that form in the tropical latitudes even more intense and more likely.
To quantify the economic damages attributable to climate change caused by a specific cyclone, Dr Theokritoff and her colleagues first need to understand the influence of human activities on the storm.
Scarcity of data means this isn’t an easy task.
Globally, only about six very damaging tropical cyclones make landfall every year. And many developing countries have less than 40 years of high-quality data, coinciding with the beginning of satellite observations in 1979.
Professor Ralf Toumi in the Data Science Institute
Professor Ralf Toumi in the Data Science Institute
To overcome this issue, Ralf Toumi, Professor of Atmospheric Physics and Director of the Grantham Institute, and research associate Dr Nathan Sparks developed the Imperial College Storm model (IRIS), which simulates millions of tropical cyclones tracks under a determined level of warming.
The model relies on past tropical cyclones’ wind speed – which is often used to measure the intensity of a storm – and climate data since 1980. It then generates ‘synthetic’ cyclone tracks for any location in the world for a counterfactual 10,000 years by perturbing the paths of historically observed storms to make them enter the studied area.
For the first time, this allows researchers to calculate how much more intense a cyclone’s wind speed is today compared with a 1.3°C cooler world without climate change, and this within 24 hours of a storm making landfall. Professor Toumi says the IRIS model was “built with economic damage in mind”, describing it as “an open version” of the type of proprietary models used by the insurance industry.
It found that climate change increased Hurricane Helene’s wind speed at landfall by about 13 miles per hour or 11%, and Milton’s by almost 11 miles per hour or 10%.
The seemingly small numbers hide devastating additional damages. In a world without climate change, a category 3 hurricane of the intensity of Milton would have made landfall as a weaker category 2 hurricane. With every increase in category, the damage caused goes up by a factor of four.
To calculate how this translates into impacts on the ground, researchers use a damage function, which shows the additional economic costs caused for every increase in the wind’s intensity. The function is calibrated for each country with the costs of past cyclones, implicitly taking into account damages caused by storm surges, rain and flooding.
The results are overlayed with a map of global assets enabling researchers to see how many billions of dollars’ worth of buildings, infrastructure and machinery have been exposed to the cyclone’s track.
The team found Milton caused direct economic damages of $50 billion, 45% of which, or $23 billion, were caused by climate change.
The methodology isn’t granular enough to provide details on the severity of the damage and indirect costs, or take into account long-term impacts. “This is a snapshot when the event happens,” explains Dr Theokritoff.
But even if the amount of damage varies, the team is confident in the fraction of the losses the method attributes to climate change – a value which remained stable “even if the losses were different by a factor of 10,” says Professor Toumi.
For Professor Joeri Rogelj, Grantham Institute Director of Research, the initiative fills a critical information gap about the impact of extreme weather events at a time when the public conversation is still focused on the disaster.
“Our working assumption is that what matters to people is what they experience most directly, and that is the damages, the excess deaths, the effects it has on people's lives,” he says.
Currently, estimates of the losses caused by extreme events don’t distinguish human activities’ share of responsibility and those that do are often published long after the event.
By rapidly publishing an estimate of the damages attributable to climate change, researchers can inform the societal conversation in a timely manner, says Professor Rogelj – a role he sees as more important than ever at a time when the scientific consensus on human-caused climate change is coming under attack.
The research is already proving valuable to help governments better understand the climate shocks coming their way.
Professor Toumi says the team is running simulations to help the government of Mozambique understand the likelihood of major tropical cyclones hitting the Southern African nation today and in a warmer world, and the impacts these could have.
Meanwhile, there is a vivid debate among scientists about the role climate damage attribution could play in informing discussions on the support vulnerable communities who are victim of climate-fuelled disasters should receive under the UN’s Loss and Damage Fund.
By quantifying the direct economic losses that occur due to climate change, impact attribution studies demonstrate the additional risks communities around the world are exposed to.
This could help inform discussions about the level of financial support a country hit by an extreme event should receive, says Professor Toumi, who sees in the team’s global open-source model a transparent way to “put flesh on that debate” by presenting a number for discussion.
However, access to loss and damage funding shouldn’t be conditional on an attribution study, warns Dr Theokritoff, citing limited data in some of the world’s most climate vulnerable regions as a major obstacle.
The team encountered these challenges firsthand when Cyclone Chido hit the French Indian Ocean territory of Mayotte, which doesn’t usually find itself on the path of such a strong tropical cyclone.
Chido obliterated the archipelago’s slums, where a large but unknown number of undocumented migrants live. Unable to capture the economic damage caused to informal settlements, the team decided not to carry out an impact attribution study, concerned it wouldn’t truly reflect the scale of the devastation.
Better capturing this contextual information is a key challenge for the researchers.
“There is a bit of a risk of focusing on the things that you can quantify and forgetting all those other things that you can't quantify”, such as early warning signals or how different groups of people might be more vulnerable to shocks, says Dr Clair Barnes, a research associate with the WWA.
“Being able to shine a spotlight on regions where climate impacts perhaps aren't reported for systemic reasons is really important. Where impacts aren’t reported, it doesn’t mean they are not happening,” she adds.
Another area where climate damage attribution is already playing a bigger role is in the courts.
In 2024, a group of elderly Swiss women won a landmark case in the European Court of Human Rights. The women, mostly in their 70s, argued that their age and gender put them at greater risk of deaths during climate-fuelled heatwaves. An impact attribution study, which quantified the number of deaths attributable to climate change in the Swiss Canton of Zürich over a 50-year period, formed part of the evidence.
Climate litigation cases are more likely to rely on slower and more precise studies with specific local data than on the rapid and global studies planned by Imperial scientists. But any advancement in the field will benefit climate litigation cases, says Dr Theokritoff. Eventually, impact attribution could be combined with another field of research concerned with attributing the contribution of different sectors or entities to climate change to demand the world’s biggest polluters pay for the damages attributable to human activities.
As the start of the next hurricane season looms, Dr Theokritoff plans to carry out rapid attribution studies for all category 3 and above tropical cyclones that make landfall around the world.
Over time, she hopes to expand the work to different types of impacts and extreme weather events, including by estimating excess mortality and reduced labour productivity during heatwaves.
But establishing methodologies that can work quickly and on a global scale is challenging. For example, estimating the number of additional deaths caused by climate change during extreme heat would require quantifying how much more intense the heat is because of climate change and then calculating the deaths caused by this additional heat. Doing so, would require accessing death counts in real-time, which mostly isn’t possible, she explains.
In addition, it isn’t sufficient for researchers to establish how many people are exposed to the heatwaves, they also need to understand how they live and how vulnerable they might be to its effects.
Yet, overcoming these difficulties opens up opportunities to carry out further research and foster new collaboration within and beyond Imperial, says Professor Rogelj.
Getting around the data gaps requires finding innovative ways to piece together the missing information. For example, estimating the extent of informal settlements could involve using satellite information, light data or the burning of biomass to identify where populations live and how vulnerable to shocks they might be – all of which opens research opportunities for students with different skillsets and academic backgrounds, he adds.
“None of this is straightforward and easy. But it provides good opportunities to develop these methods and to make the very vulnerable visible.”
Dr Clair Barnes and Dr Emily Theokritoff
Dr Clair Barnes and Dr Emily Theokritoff
Professor Joeri Rogelj and Dr Gary Konstantinoudis
Professor Joeri Rogelj and Dr Gary Konstantinoudis
Built with Shorthand