EGU22-10642, updated on 28 Mar 2022
EGU General Assembly 2022
© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.

Identification of European heatwave families 

Julia Hellmig, Felix Strnad, and Bedartha Goswami
Julia Hellmig et al.
  • Eberhard Karls University Tübingen, Cluster of Excellence Machine Learning, Machine Learning in Climate Science, Tübingen, Germany (

Mainly caused by anthropogenic climate change occurring heatwaves have become more frequent and extreme throughout the 21th century. Summer heatwaves over Europe are mainly caused by positive phases of the North Atlantic Oscillation (NAO) and jet stream anomalies, subsequently causing atmospheric blocking over different parts of Europe. With this work we aim to define families of European heatwaves caused by different atmospheric regimes. In the long run this could help predicting European heatwaves and their length, intensity and spatial extend. To identify European heatwaves and their spatial extend we use the graph framework DeepGraphs. Within this framework every extreme heat day isconsidered a node and a heatwave is defined as the union of all nearest neighbour nodes (which are connected by edges). 

Two clustering steps are applied to cluster the heatwave into families depending on their length, season and spatial extend. 

Our results reveal a promising way to classify European heatwaves based on their atmospheric cause which could help forecasting heatwaves in the future.

How to cite: Hellmig, J., Strnad, F., and Goswami, B.: Identification of European heatwave families , EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-10642,, 2022.