Most climate forecasts are based on averages of multiple climate models and global trends. But this may result in certain climate impacts being underestimated, as a study now demonstrates. In it, climate researchers have determined that the weather extremes that were actually only predicted for three to four degrees of warming compared to pre-industrial values could occur for some regions and models at just two degrees of warming. This applies, for example, to droughts in important grain-growing areas, heavy rain in metropolitan areas or fire-promoting weather in forests. The current forecasts could therefore substantially underestimate the risk of such climate impacts – but the uncertainties are extremely large.
Typically, forecasts for climate impacts and the risk of upcoming weather extremes are created through simulations of several different climate models – this helps to compensate for the idiosyncrasies of the models. The results of the different models are averaged for the final result. But this also means that outliers on both sides no longer appear in this final result. The problem is that it is not yet clear which of these climate models most realistically reflect future developments. “There are considerable uncertainties about which climatic drivers come into play at different degrees of warming,” explain Emanuele Bevacqua from the Helmholtz Center for Environmental Research (UFZ) in Leipzig and his colleagues. This could mean that some of the models declared as outliers may be more accurate than those in the middle with their forecasts. If you ignore this wide range of forecast results, you may be overlooking climate risks, especially at the regional level.
Large range for the worse
“In the spirit of responsible risk assessment, we should therefore look beyond the most likely developments and also consider extreme scenarios that could have serious social or ecological consequences,” says Bevacqua. For their study, he and his team therefore selected the forecasts for three areas that have particular social and ecological significance. These include heavy rainfall in densely populated regions, droughts in important agricultural areas and extreme fire weather conditions in forests. For their simulations, they used the CMIP6 climate models – the suite of models on which the Intergovernmental Panel on Climate Change (IPCC) reports are based. For each of the three weather extremes, the researchers identified the sector-specific climate factors and then ran climate projections assuming progressive climate change up to a warming of two degrees.
The result was a wide range of projections for each of the three sub-scenarios considered. The crucial thing is that for all three scenarios, individual forecasts were significantly above or below the average values normally considered. At the upper extreme, it was shown that, according to some models, climate consequences in the form of forest fires, drought in growing areas or heavy rain in metropolitan areas could be more extreme at just two degrees of warming than previously predicted for three to four degrees. In concrete terms, this means, for example, that the important growing regions for corn, wheat, soy and rice could experience up to 50 percent more droughts at just two degrees of warming. “Ten of the 42 models examined deliver results at two degrees that are significantly above the model average at four degrees of warming,” reports Bevacqua. If these models are correct, the consequences for food security, global supply chains and international markets would be significant. Some models for heavy rain in metropolitan areas or fire weather in forests produced similar worst-case results.
Extreme insecurities
“Our results do not mean that a two-degree warming would be as serious overall as a significantly greater warming,” emphasizes co-author Jakob Zscheischler from the UFZ. “Rather, they show that extreme impacts can occur in particularly vulnerable or socially important sectors even with moderate warming of two degrees.” Moderate global warming is therefore no guarantee of moderate effects. “Because the projections are subject to large uncertainties, extreme climate developments are possible even with global warming of two degrees and are often underestimated when the focus is on model averages,” says Bevacqua. “Focusing on average values can contribute to a false sense of security.”
Climate researchers who were not involved in the study also see it similarly: “The study shows us once again how uncertain we are about what a certain global warming ultimately means. Be it changes in heavy rain in populated places, droughts in agriculturally important regions or fire weather in forest areas: In these and other important areas it is still far from clear what global warming of two degrees Celsius actually means,” comments Helge Gößling from the Alfred Wegener Institute, Helmholtz Center for Polar and Marine Research (AWI) in Bremerhaven. However, the study should not be interpreted in the sense of ‘everything will get much worse’. “Rather, it shows impressively that we still have glaring gaps in our knowledge of the expected climate impacts that should urgently be closed.” At the same time, the results also provide another reason why climate change should be limited to warming below two degrees, if possible: As long as we don’t know whether the worst case scenario will occur, we should err on the side of caution.
Source: Emanuele Bevacqua (Helmholtz Center for Environmental Research – UFZ, Leipzig) et al., Nature, doi: 10.1038/s41586-026-10237-9