Climatology and Mechanisms of North American Heat Waves

Heat waves (HWs) are the leading cause of weather-related mortality in the United States, and are expected to increase in frequency and intensity with climate change. This study seeks to better understand mechanisms for North American HWs using synoptic analysis, self-organizing maps, and Lagrangian back trajectories.

To do this, we first conduct a climatology of North American HWs. Extreme heat events are identified over 1979-2024 with the criteria that they are extreme in temperature, spatiotemporally contiguous, and unusual for a given region. We include nightly minimum temperatures and humidity in our identification in order to exclude events which are less likely to impact human health. Examination of these HWs reveals variability in characteristics from region to region, and an upward trend in HW frequency and area over time.

We then use self-organizing maps (SOMs), a machine learning technique, as a means to uncover variability in mid-tropospheric geopotential height patterns driving heat waves. SOMs are trained on event-relative 500-hPa geopotential height anomalies. Preliminary results show substantial variability in patterns leading to HWs, even within a single region. We then select highly representative cases from each SOM cluster for additional analysis.

Lagrangian back trajectories serve as a useful tool for identifying heat wave source air masses, and for quantifying heating mechanisms (e.g. subsidence, advection, radiation, condensation). We calculate five-day back trajectories using the LAGRANTO software package for each representative case to gain insight into the dynamic and thermodynamic mechanisms responsible for the heat event. Early trajectory results show a variety of heating mechanisms, for instance, some trajectories conserve theta, while others exhibit a strong diabatic heating signature.