EGU General Assembly 2020
© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

Heatwaves over Europe: Identification and connection to large-scale circulation

Emmanuel Rouges1,2, Laura Ferranti1, Holger Kantz2, and Florian Pappenberger1
Emmanuel Rouges et al.
  • 1ECMWF, Forecast Department, Reading, United Kingdom of Great Britain and Northern Ireland (
  • 2Technische Universität Dresden, Dresden, Germany

                Heat waves have important impacts on society and our environment. In Europe for instance, the summer of 2003 caused upwards of 40000 fatalities. They also impact the crop production, ecosystems, and infrastructures. In a warming climate, heat wave intensity and frequency are likely to increase with potentially more dramatic consequences.

                Considering this, it is crucial to forecast such extreme events and therefore gain a better understanding of their triggering processes. The determination of these processes requires to identify heat wave patterns (timing and location) together with the correlated large-scale circulation patterns. This will enable to devise early warning systems, that could help mitigate the impact.

                This work is part of an ongoing PhD project focusing on improving the forecast of heat waves at sub-seasonal time scale. The main objectives are to evaluate the link between large scale weather patterns and severe warm events over Europe and measure current level of predictive skill. The first part will focus on defining an objective criteria to identify heat wave events in the ERA5 reanalaysis dataset from ECMWF. The identification of heat waves depends on three main criteria: temperature threshold, spatial and temporal extension. Meaning that the temperature should exceed a defined threshold over a large enough region and for a long enough period. We will consider daily means as well as maximum and minimum values of 2m temperature. We will identify the circulation patterns (persistent high pressure systems) associated with heat wave events and analyse the key differences with persistent high pressure systems that are not associated with heat waves.

                This work is part of the Climate Advanced Forecasting of sub-seasonal Extremes (CAFE) project, funded by the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grand agreement No 813844.

How to cite: Rouges, E., Ferranti, L., Kantz, H., and Pappenberger, F.: Heatwaves over Europe: Identification and connection to large-scale circulation, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-5619,, 2020


Display file

Comments on the display

AC: Author Comment | CC: Community Comment | Report abuse

displays version 1 – uploaded on 01 May 2020
  • CC1: Comment on EGU2020-5619, Pedro M. Sousa, 04 May 2020

    Dear Emmanuel and co-authors,

    Very interesting presentation. We've been looking closely at blocking/ridge patterns and their links with extreme temperature events in Europe, and working on a combined catalogue which could be used for operational purposes. So this is very interesting regarding applications at synoptic time-scales.

    Some brief comments/questions:

    1) In my point of view, the pattern for NS resembles more a Rex type blocking, as negative Z500 anomalies are found in almost the entire longitude span "below" the anticyclone.
    2) How are you thinking of explicitly verifying the predictability, based on the Hovmöller plots shown here?
    3) Are you thinking of designing some "simple" statistical models to try to produce such forecasts? Based on simple indicators like Weather Types and soil-moisture?

    Best regards,
    Pedro Sousa

    • AC1: Reply to CC1, Emmanuel Rouges, 04 May 2020

      Dear Pedro,


      Thank you very much for your interest and comments.

      1) I was not aware of this ‘Rex’ type blocking and after looking quickly, I think I would still categorise the NS blocking as a ‘omega’ type blocking. The composite is an average over all events and days for this category and some features are not representative of all events. Looking at the individual events (not shown here), the blocking type for most cases is the ‘omega’ blocking. But thank you for making me aware of this blocking type and I will look closer to differenciate them better.

      2) The Hovmöller diagrams are mostly here to show that we have common large-scale features to all cases of one category that happen before the occurrence of the event, making them precursors. These precursors then become a source of predictability for the S2S time range. I do not yet intend to use the diagrams to explicitly the predictability.

      3) The aim of the thesis is to investigate the forecasting skill of S2S models for heat waves and their precursors. It will not involve designing some statistical model but probably some post-processing procedure to identify and predict favourable conditions for heat waves.


      Let me know if these answers have clarified my work for you, otherwise you can contact me by e-mail.


      Best regards,

      Emmanuel Rouges

  • CC2: Comment on EGU2020-5619, Philipp Zschenderlein, 04 May 2020

    Hi Emmanuel,

    thanks for your presentation. We are looking at the dynamics of European heat waves and your project is very interesting for us.

    I have some specific questions:

    (1) How do you quantify the slow establishment of Z500-ridges? Do you have an idea how long the ridge building typically takes?

    (2) It's interesting to see that the amplitude of Z500 ridges is very high in Northern Europe, and flatter for Central Europe (comparable to Sousa et a., 2018). Have you also looked at southern Europe?

    (3) Have you looked at ridges that are not connected to surface heat waves? Do you see differences in the establishment of these ridges?




    • AC2: Reply to CC2, Emmanuel Rouges, 04 May 2020

      Hi Philipp,

      I am happy that you are interested in my work, especially since I was very interested by your work in your last paper (A Lagrangian analysis of upper-tropospheric anticyclones associated with heat waves in Europe).


      (1) For now, we have not quantified the slow establishment of Z500-ridges, this is on observation looking at the individual events and the average Hovmöller diagrams. The precise quantification is part of my next steps

      (2) I have not yet, but will be interesting to see.

      (3)This is also part of the next investigation, identifying the observed features across the ERA5 period and looking at the ‘false alarms’. The aim being to understand why these ridges did not lead to heat waves (what is different in these ridges, are other key ingredients necessary e.g. low soil moisture content…)


      I hope this answered your questions, but be free to ask more questions by e-mail.

      Best regards,