Global spatiotemporal patterns of heatwaves: a re-evaluation

Heatwaves are among the most impactful climate extremes, yet their detection and quantification depend strongly on methodological choices, particularly the definition of the baseline climate. In this study, we conduct a global analysis of atmospheric heatwaves from 1981 to 2023 using ERA5-Land reanalysis data to evaluate how baseline selection influences the characterization of heatwave frequency, duration, and intensity. We compare two commonly used approaches: a traditional fixed baseline (1951-1980) and a running baseline defined by the preceding 30-year climatology. In addition, we introduce a ratio-based intensity metric that normalizes temperature anomalies by local background variability to better account for differences in climatic regimes. Results show that baseline choice substantially affects the apparent magnitude of heatwave trends. Using a fixed baseline, global heatwave frequency, duration, and intensity exhibit strong and widespread increasing trends. In contrast, the running baseline attenuates these trends by adjusting for long-term warming, highlighting deviations relative to evolving climatic conditions rather than historical climatology. Despite these differences in trend magnitude, the overall spatial distribution and frequency patterns of heatwaves remain broadly consistent between the two approaches. When heatwave intensity is expressed relative to local variability, the pronounced latitudinal gradient observed with traditional absolute anomalies becomes less distinct, suggesting that relative thermal exposure may be more evenly distributed across climatic zones than previously inferred. Running baselines and variability-normalized metrics provide complementary perspectives for interpreting extremes under a non-stationary climate. These findings highlight the importance of methodological transparency when comparing heatwave studies and assessing ecological and societal risks associated with extreme heat.