Mapped: How climate change affects extreme weather around the world

Extremes across the world

In total, the map (above) contains 967 extremes from 819 studies. This is a combination of individual events, such as the “exceptional megadrought” on Easter Island over 2010-23, and trends of how extremes are changing, such as the 2024 study on increasing compound “hot-dry” events in China.

Easter Island 'megadrought', 2010-23

Case 643

Event type

Drought

Finding

More severe or more likely to occur

We attribute the current megadrought on Rapa Nui as being part of a long-term trend driven by greenhouse-gas forcing.

Geophysical Research Letters (2025) Markovitz, E. et al., 2025: Contemporary Megadrought on Easter Island (Rapa Nui) Since 2010, Geophysical Research Letters, doi:10.1029/2025GL115880

China hot-dry events, 1979-2014

Case 607

Event type

Compound

Finding

More severe or more likely to occur

"Over the past 40 years, anthropogenic activities have increased hot-dry events by 2.34 timess."

Communications Earth & Environment (2024) Yao H. et al., 2024: Changes caused by human activities in the high health-risk hot-dry and hot-wet events in China, Communications Earth & Environment, doi:10.1038/s43247-024-01625-y

Where a single study covers multiple events or locations, these have been separated out into individual entries on the map (where possible).

Such is the sheer quantity of attribution studies, the new map displays them in regional summaries, rather than individual events or trends. (The regions follow those used by the UN, except the UN’s “Europe and Northern America” region has been divided into two.)

By default, the map displays the data by attribution result.

The circles in each region indicate how many events and trends have been made more (red) or less (blue) severe or likely by climate change.

Yellow indicates extremes where no human influence was found, while grey shows studies that were inconclusive.

To explore the data for a specific region, click the relevant circles on the map. This reveals a summary page and a link to explore all the studies related to that part of the world.

Across all cases, 74% were found to have been made more likely or severe because of climate change. More than a third of these are heat extremes, which are generally the most straightforward events to link to a warming world.

Okavango basin floods, 2009-11

Case 076

Event type

Rain & flooding

Finding

Decrease, less severe or less likely to occur

"Anthropogenic climate change has reduced probability of high floods in the Okavango. Reduction is due to increased evaporation; rainfall changes are minor."

Journal of Hydrology (2014) Wolski, P. et al., 2014: Attribution of floods in the Okavango basin, Southern Africa, Journal of Hydrology, Vol 511, 16 April 2014, Pages 350-358

Around 9% of the events and trends in the map were made less likely or severe by climate change. Unsurprisingly, this category is dominated by blizzards and cold extremes, but there are also cases where climate change has lessened the chances of other extremes – such as the flooding in the Okavango basin in southern Africa in 2009-11.

The remaining cases are extremes where scientists either identified no human influence (8%) or where the findings were inconclusive (7%).

(It should be noted that these figures are not representative of all extreme weather events as only a small fraction have been subject to an attribution study. In addition, some of the events included in the map have been the subject of more than one study.)

The map can also be viewed by event type, which groups the data into seven categories: heat (red), rain and flooding (navy blue), drought (yellow), storm (purple), cold, snow and ice (blue) and other (grey).

To delve into the individual extremes, click “explore all cases” in the map legend or choose a specific country from the dropdown list. A searchable table is also included at the end of this article.

Evolving methods

Following the first extreme event attribution study in 2004, the research field has rapidly gained momentum – with many more scientists and institutions carrying out studies and new methods being developed.

And what started as a trickle of studies has turned into a flood. In the 10 years following that first scientific paper, around 50 more were published. In the 10 years after that, the number has risen tenfold by more than 500.

There are various ways of carrying out an attribution study, but scientists commonly use climate models to simulate an extreme event in the current climate and compare them with idealised model runs of that event in a world without human-caused warming. The difference between the two sets of simulations indicates how the likelihood or severity of that extreme event has changed.

The evolution of attribution methods over the past two decades is explored in detail in an accompanying Carbon Brief article.

The chart below reveals that the most-studied extremes are related to heat (27%) and rainfall and flooding (23%), which together account for more than half of the events and trends in the map. The next-largest group is for drought (13%), followed by storms (9%) and cold, snow and ice (7%).

One of the most significant developments seen in attribution science is the advent of “rapid” studies.

In 2015, the World Weather Attribution (WWA) initiative was founded, which streamlined the attribution process in order to produce results within days of an extreme event occurring.

The team uses a standard, peer-reviewed methodology for their analysis, but does not routinely publish the results in formal journals – instead publishing them directly on the WWA website as soon as the analysis is complete.

This has allowed scientists to answer the question of how climate change contributed to an extreme event in the immediate aftermath, rather than months later.

For example, extreme heat in May and June killed more than 100 people in Mexico. Within weeks, WWA was able to show that human-caused climate change had made the event’s maximum temperatures around 1.4C hotter and about 35 times more likely.

Mexico extreme heat, 2024

Rapid study. Case 554

Event type

Heat

Finding

More severe or more likely to occur

"Human-induced warming from burning fossil fuels made the 5-day maximum temperature event about 1.4 degrees hotter and about 35 times more likely. For nighttime temperatures this is about 1.6C hotter and about 200 times more likely."

World Weather Attribution (rapid study) (2024) Pinto, I. et al, 2024: Extreme heat killing more than 100 people in Mexico hotter and much more likely due to climate change, World Weather Attribution

The Carbon Brief map includes WWA studies as well as those from selected other groups, including the Grantham Institute at Imperial College London and the UK Met Office. (See methodology at the end of this article for more details.)

Another recent advance is “impact attribution”, which quantifies the social, economic and/or ecological impacts arising from the influence of climate change on extreme weather.

Carbon Brief has a specific category for all of these studies. These include, for example, research showing that human-caused warming was responsible for more than half of the deaths during Europe’s record-breaking hot summer in 2022, and a study attributing almost one-third of total landslide damages in central China over 2011-20 to climate change.

Mortality due to 2022 European summer heat

Case 807

Event type

Impact

Finding

More severe or more likely to occur

By combining epidemiological models with detection and attribution techniques, we attribute half of this mortality burden (~56%) to anthropogenic warming.

npj climate and atmospheric science (2024) Beck, T. et al., 2024: Mortality burden attributed to anthropogenic warming during Europe’s 2022 record-breaking summer, npj climate and atmospheric science, doi: 10.1038/s41612-024-00783-2

Rainfall-induced landslides in China over 2011-20

Case 696

Event type

Impact

Finding

More severe or more likely to occur

We find that climate change contributed to 32.8% of the total landslide damages in Jiangxi Province over the past decade.

International Journal of Disaster Risk Reduction (2025) Fang, Z. et al., 2025: Climate change has increased rainfall-induced landslide damages in central China, International Journal of Disaster Risk Reduction, doi:10.1016/j.ijdrr.2025.105320

Uneven spread

Even a cursory look at the Carbon Brief map reveals the uneven spread of studied extremes across the world, with the vast majority in the global north.

The cases included in the map are dominated by extremes in Europe (22%), eastern and south-east Asia (22%) and northern America (18%). In contrast, relatively few of the studied extremes are in central and southern Asia (6%), Oceania (1%) and northern Africa and western Asia (1%).

There are a number of reasons for this, including a lack of weather data and monitoring of extremes in many developing countries. Another factor is that scientists and their institutions that conduct attribution research are often themselves based in global-north countries.

This imbalance is something that many attribution scientists are trying to address, putting a greater focus on extremes in countries that are often overlooked.

What is also noticeable in this latest iteration of the map is the large number of studies examining extremes in China, following record-breaking heatwaves, severe drought and deadly rainfall events in recent years.

In total, 162 extremes and trends in China have been the subject of an attribution study – making up 17% of all the cases included in the map. Around 80% of them have been published in the past five years.

Use the searchable table below to see all extremes and trends for a specific country, and/or use the dropdown lists to select specific regions or types of extremes.