Comeau et al. (2025) From mild to extreme heatwaves: Examining trends in North America

Identification

• Journal: Weather and Climate Extremes
• Year: 2025
• Date: 2025-11-16
• Authors: Élise Comeau, Alejandro Di Luca, Megan C. Kirchmeier‐Young
• DOI: 10.1016/j.wace.2025.100831

Research Groups

• Département des sciences de la Terre et de l’atmosphère, Centre pour l’étude et la simulation du climat à l’échelle régionale (ESCER), Université du Québec à Montréal, Montreal, Canada
• Climate Research Division, Environment and Climate Change Canada, Toronto, Canada

Short Summary

This study quantifies the intensity and duration trends of North American heatwaves from 1940 to 2019, introducing a novel severity metric to differentiate between mild, moderate, and extreme events. It finds that while heatwave temperatures have increased due to climate warming, the temperature anomalies relative to a changing threshold have remained largely stable, with higher severity heatwaves exhibiting more spatially coherent trends influenced by changes in both the mean and standard deviation of the temperature distribution.

Objective

• To quantify recent intensity and duration trends of North American heatwaves (1940–2019) according to their severity, a novel metric measuring heatwave intensity relative to other local contemporaneous heatwaves.
• To investigate how past trends of milder heatwaves may differ from those of more extreme heatwaves.
• To understand how changes in the distribution of daily maximum temperature (TX) over time impact heatwave intensity trends.

Study Configuration

• Spatial Scale: North America (land between 220° and 315° E longitude and 15° and 65° N latitude).
• Temporal Scale: 1940–2019 (80 years) for historical analysis; heatwave severity classified using consecutive 10-year intervals.

Methodology and Data

• Models used:
  • Numerical experiments: Simulated daily maximum temperature (TX) values using a random Gaussian distribution, varying the annual mean and standard deviation.
  • Statistical methods: Theil-Sen slope estimator for trend calculation, and a modified Mann-Kendall test (Hamed and Rao) for statistical significance (95% confidence level).
  • Python packages: statsmodels (quantile regression), Climate Data Operators (CDO) for TX calculation, pyMannKendall for trend tests.
• Data sources:
  • ERA5 reanalysis (European Centre for Medium-Range Weather Forecasts - ECMWF) global gridded dataset.
  • Hourly 2-meter air temperature (T2m) from ERA5, processed to derive daily maximum temperature (TX).

Main Results

• Widespread Temperature Increases: The 90th percentile of daily maximum temperature (TX90) increased across most of North America for every season between 1940 and 2019, indicating a rise in heatwave temperatures due to climate warming. Spring showed the highest mean trend (0.321 °C⋅decade⁻¹), while autumn had the lowest (0.160 °C⋅decade⁻¹).
• Stable Temperature Anomalies: Heatwave temperature anomalies (TXanom = TX - TX90) remained relatively stable over the 80-year period, with continent-wide mean relative intensity and duration trends typically up to 1.5 %⋅decade⁻¹.
• Severity Influences Trend Coherence: Higher heatwave severity is associated with less noisy intensity trends and greater spatial coherence. More severe heatwaves show larger increases or decreases in temperature anomalies compared to less severe ones in regions with positive or negative trends, respectively.
• Seasonal Variability: Spring exhibited proportionately larger increases in temperature anomalies compared to other seasons, though this distinction was less robust in the 1970–2019 period. Summer intensity and duration trends showed reduced spatial coherence.
• Role of TX Distribution Changes: Numerical experiments demonstrated that an increasing mean of the TX distribution alone results in constant intensity trends across all severities. However, when both the mean and standard deviation of TX increase, higher severities show greater increases in their temperature anomalies, providing a plausible explanation for observed severity-differentiated trends.

Contributions

• Introduction and application of a novel heatwave severity metric, which categorizes heatwaves as mild, moderate, or extreme based on their intensity relative to other local contemporaneous heatwaves.
• First study to analyze historical heatwave intensity and duration trends in North America by this severity classification, revealing differentiated trends.
• Utilizes a non-stationary 90th percentile threshold for heatwave definition, which dynamically accounts for climate warming and seasonal effects.
• Provides numerical experiments to elucidate the underlying mechanisms (changes in mean and standard deviation of TX distribution) that drive severity-dependent heatwave intensity trends.
• Highlights the importance of considering heatwave severity for a more nuanced understanding of heatwave intensity trends and their implications.

Funding

• Fond de recherche du Québec–Nature et technologies (FRQNT) (grant 324877)
• Natural Sciences and Engineering Research Council of Canada (NSERC) (grant RGPIN-2020-05631)

Citation

@article{Comeau2025From,
  author = {Comeau, Élise and Luca, Alejandro Di and Kirchmeier‐Young, Megan C.},
  title = {From mild to extreme heatwaves: Examining trends in North America},
  journal = {Weather and Climate Extremes},
  year = {2025},
  doi = {10.1016/j.wace.2025.100831},
  url = {https://doi.org/10.1016/j.wace.2025.100831}
}