In recent days, a series of studies and analyses has highlighted a fact that may seem implausible to some: the three hottest summers on record have been 2023, 2024, and 2025, and the extreme heat linked to climate change has contributed to thousands of deaths in European cities.
At the same time, it's true that in some parts of Europe, the summer recorded more frequent rains than usual—and someone might argue that it "didn't feel hot at all."
Let's clarify: what do the data say, how were they obtained, and even considering the limitations of statistical collection, we must accept that they are more reliable than those who say, "Well, it rained all summer at my place!"
Large-scale climate observations (Copernicus, WMO, and international datasets) show that 2024 was the hottest ever recorded and that the 2023–2025 period marks peaks in seasonal temperatures, especially in the Northern Hemisphere, among those with reliable data. These datasets measure average temperatures over vast areas and periods, capturing global/continental trends that—though it may seem obvious—do not always coincide with local perception.
Attribution: How Human Impact on Climate Is Determined
"Attribution" studies compare observed data with models simulating a world where human emissions have not influenced temperature increases. By applying these comparisons to historical temperature series and daily mortality data, researchers estimate the fraction of heatwaves (and related deaths) made more likely or frequent by human-induced warming. Recent analyses of 854 European cities estimate that in the most recent warm season, about 60–70% of observed heat-related deaths are attributable to human-induced warming.
Why Your Possibly Rainy Summer Doesn't Contradict These Findings
Different Spatial Scale
Climate datasets and mortality studies operate on a continental scale or across hundreds of cities: they capture aggregated trends. A wetter summer in an Alpine valley doesn't negate prolonged heatwaves in plains and coasts.
Meteorological Variability vs. Climate Trend
Climate change increases the probability and intensity of extremes but doesn't eliminate natural variability: there will always be intense rains, storms, and cooler regions; these are inevitable components of the new "disturbed climate."
Amplified Local Impacts
Heat-related mortality is often linked to socio-demographic factors (average population age, building quality, access to air conditioning), so two cities with the same temperature can have very different outcomes.
Methodological Limitations—Why Interpret the Numbers Cautiously
Data Coverage: The cited studies rely on city networks with complete data; rural areas or countries with weak records may be underrepresented, leading to underestimates or geographic biases.
Model Assumptions: Attribution analyses require assumptions about pre-industrial levels, emission scenarios, and human behavior: each step introduces uncertainty.
Temporal Definitions: Discussing "exceeding 1.5°C" can refer to a single average year (as reported in 2024) or multi-year periods; the political and scientific significance changes depending on the reference.
Methodological Limitations Don't Mean Distrust in the Data
Every scientific analysis carries margins of uncertainty: heat and mortality studies are based on available data (stronger in some areas, weaker in others), models simulating counterfactual scenarios, and temporal definitions that may vary. Recognizing this is correct. But it would be wrong to conclude that, because of this, the results are unreliable. On the contrary: these works are conducted with methodological rigor, peer-reviewed, and based on processing vast amounts of data from meteorology, epidemiology, and social sciences. They are tools that, year after year, become more precise thanks to improvements in historical series, climate models, and statistical techniques.
Ultimately, highlighting the limitations serves to interpret the results correctly, not to question their validity: science, with its methods, remains our most reliable compass to navigate complex and dynamic phenomena like climate. Distrusting it in favor of "personal perception" is a shortcut we cannot afford.
Main sources |
Scientific American: Andrea Thompson, Climate Change Fuels Record Summer Heat, Killing Thousands, 17 settembre 2025. Scientific American |
Copernicus Climate Change Service: Rapporto Europe’s Warmest Year on Record—Striking Climate Contrasts, 17 aprile 2025. copernicus.eu |
World Meteorological Organization (WMO): Rapporto European State of the Climate 2024, 15 aprile 2025. World Meteorological Organization |
ISGlobal: Studio 62.700 heat-related deaths in summer 2024, 22 settembre 2025. ISGLOBAL |
Imperial College London: Studio Summer heat deaths in 854 European cities more than tripled due to climate change, 17 settembre 2025. Imperial College London |
Nature Medicine: Studio Heat-related mortality in Europe during the summer of 2022, 2023. Nature |
Recent numbers—including estimates pointing to tens of thousands of heat-related deaths in Europe between 2022 and 2024—converge on one conclusion: extreme heat is already a concrete threat to public health and European infrastructure. Adaptation policies (heat shelters, early warning systems, greener cities, social infrastructure) are just as crucial as mitigation efforts that reduce future emissions and act on mechanisms to remove CO₂ from the atmosphere.
The misalignment between local experience (“it rained here”) and macro trends (“summers have become hotter and deadlier”) is not a contradiction: it reflects the coexistence of local meteorological variability and systemic changes in the global climate.
Data do not cancel out personal experience, but they frame it: they outline scenarios that require collective measures—urban adaptation, protection of the most vulnerable, and emissions reduction—as well as everyday choices that include land management and the restoration of ecosystem services. Within this framework, planting and protecting trees is one of many local actions that contribute both to adaptation (shading, local cooling) and climate mitigation. But it is not a single solution: a combination of policies, infrastructure, and data-informed behaviors is needed.