Variability of Hot Days in the Middle Latitudes of Europe between 1973 and 2024

Hot days gain great attention around the world for their wide impacts on water resources, agriculture, society productivity, biosystems, and public health. We explore daily data from weather stations, the global surface summary of day product produced by the National Centers for Environmental Information of the United States, to determine changes in the monthly frequency of hot days (maximum daily temperature above 35 degrees Celsius) in Western, Central, and Eastern Europe. The monthly percentages show increasing trends in June, July, and August between 1973 and 2024. They are positively correlated with the monthly average geopotential height at 500hPa: Correlation coefficients computed with monthly ERA5 reanalysis data over the three regions lie between 0.54 and 0.78 (p<0.01). We further investigate the impact of anthropogenic global warming and internal climate modes such as the Pacific Decadal Oscillation (PDO) and El Niño-Southern Oscillation (ENSO) using geopotential height at 500 hPa and the surface air temperature (SAT) from ERA5. We draw on results for the three regions based on a decomposition method, Cyclo-Stationary Empirical Orthogonal Functions. We illustrate the method for two exemplary time points: July 1976, for which the hot days percentage was the least (1.3%), and July 2024, for which the percentage was the most (16.3%) in the considered areas. As expected, the anthropogenic global warming contributed to an increase in SAT between the two example months. By contrast, PDO statistically contributed to slightly lower SAT in July 2024 but slightly higher SAT in July 1976. Similarly, the ENSO mode played a small positive (negative) role in SAT in the West, Central Europe, but a slightly negative (positive) role in Eastern Europe in July 1976 (2024). In all three modes, the geopotential height positive and negative anomalies are consistent with those in SAT. Positive anomalies in geopotential height are usually accompanied by subsidence and strong solar irradiance at the surface, and therefore favor SAT increase, and vice versa. Regressing the anthropogenic global warming mode on SAT in Europe for the five July months with the most (in 2024, 2007, 2015, 2012, 2023) and the least hot days percentages (in 1976, 1992, 1979, 1986, 1989), clearly shows a positive impact for the former and a negative contribution for the latter, except 1989. We test the robustness of our results by comparison to a second method, Low-Frequency Component Analysis. Our results will enhance the understanding of the influence of forced and internal climate variability, specifically, modes of variability, on the frequency of hot days in the mid-latitudes.