Global analysis of the vertical temperature anomaly structure of surface heat extremes

Despite extensive research on processes leading to surface heat extremes, so far only a few studies have investigated the vertical temperature profile during heat extremes.
This study aims to fill this gap by providing a global analysis of the vertical temperature profile during heat extremes using ERA5 data during the period from 1951 to 2021. We systematically characterize the vertical temperature anomaly profiles, i.e., the deviation from the climatological profiles, during heat extremes, which were defined as the maximum hourly 2-m temperature of the year (so-called TXx events). At every grid point, the temperature anomaly profiles were determined for all 71 TXx events in the considered period, and these profiles were normalized first with the climatological temperature variance at the respective level and second with the normalized surface temperature anomaly. A median profile for each grid point was derived from these normalized temperature anomaly profiles. Then, these median temperature anomaly profiles across the globe were clustered with a k-means approach, which reveals a set of distinct vertical profiles and their regions of occurrence. In tropical regions, extreme heat events have positive temperature anomalies confined to the lower troposphere, whereas positive temperature anomalies associated with heat extremes in extratropical regions tend to extend throughout the troposphere.
In particular, heat extremes in continental mid-latitude regions feature a common median normalized temperature anomaly profile that extends throughout the troposphere. Normalized temperature anomaly profiles during recent record-breaking heatwaves have the same shape as the climatological heat extreme cluster, which indicates that temperature anomaly profiles during the most extreme heatwaves have a similar vertical structure as during average TXx events. Our approach also allows investigating the temporal evolution of these clusters, which provides insights into the three-dimensional processes driving heat extremes, e.g., whether heat extremes typically develop bottom-up or top-down. By advancing our understanding of their vertical structure, this study yields novel information about the processes leading to surface heat extremes and determining their intensity.