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

Storylines of the 2018 Northern Hemisphere heat wave at pre-industrial and higher global warming levels

Kathrin Wehrli, Mathias Hauser, and Sonia I. Seneviratne
Kathrin Wehrli et al.
  • Institute for Atmospheric and Climate Science, ETH Zurich, Zurich, Switzerland (

The 2018 summer was unusually hot in large areas of the Northern Hemisphere and simultaneous heat waves on three continents led to major impacts to agriculture and society. The event was driven by the anomalous atmospheric circulation pattern during that summer and it was only possible in a climate with global warming. There are indications that in a future, warmer climate similar events might occur regularly, affecting major ‘breadbasket’ regions of the Northern Hemisphere.

This study aims to understand the role of climate change for driving the intensity of the 2018 summer and to explore the sensitivity to changing warming levels. Model simulations are performed using the Community Earth System Model to investigate storylines for the extreme 2018 summer given the observed atmospheric large-scale circulation but different levels of background global warming: no human imprint, the 2018 conditions, and different mean global warming levels (1.5°C, 2°C, 3°C, and 4°C). The storylines explore the consequences of the event in an alternative warmer or colder world and thus help to increase our understanding of the drivers involved. The results reveal a strong contribution by the present-day level of global warming and provide an outlook to similar events in a possible future climate.

How to cite: Wehrli, K., Hauser, M., and Seneviratne, S. I.: Storylines of the 2018 Northern Hemisphere heat wave at pre-industrial and higher global warming levels, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-13802,, 2020

Comments on the presentation

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

Presentation version 2 – uploaded on 05 May 2020
Additional appendix slide, some structuring.
  • CC1: Comment on EGU2020-13802, Linda van Garderen, 05 May 2020

    Hi Kathrin,

    Thank you very  much for your presentation. I like it a lot as I am working on something similar.

    I have two method based questions.

    The first would be about the SIC. You mention they are derived from SST anomalies. Could you perhaps explain in a bit more detail how you did that for the prehistoric SIC? How did you make sure the numbers made sense? Did you use a linear approach?

    My second questions is concerning the nudging. Are you nudging at all wavelengths? Or only the larger scale?




    • AC1: Reply to CC1, Kathrin Wehrli, 05 May 2020

      Hi Linda!

      Thank you for the question! To generate SIC monthly SST and sea ice anomalies are computed for the years 1996-2015 from the climatology of the same years (from NOAA OIv2). A linear regression is then fitted to the anomalies for each month-of-the-year, a longitude-band and for both hemispheres separately. We only consider grid cells that undergo a change of sea ice fraction of over 50% for the month in consideration. The key is that the grid cells need to undergo enough variability of both SIC and SST (in hindsight I would probably use a SST criterion). The slope and intercept from the regression are smoothed zonally. In the natural case (1861-1880 conditions) I have a negative deltaSST (derived from CMIP5) from which I get a deltaSIC using the regression. DeltaSST and deltaSIC are in my case added to the observed ocean of 2018. To make sure the numbers make sense the new set of SSTs and sea ice are then adjusted according to the constraint of Hurrell et al. (2008),, which ensures that:

      • (i) sea ice fraction is 100% at -1.8°C, and SSTs do not get colder than that,
      • (ii) there is no sea ice at water temperatures warmer than 4.97°C, and
      • (iii) that within this temperature range the maximum sea ice fraction is limited by a temperature-dependent function.

      To your second question: We didn’t run the model on a spectral grid, therefore all wavelengths were nudged. However, a height-dependent nudging coefficient was used, which is zero at the ground (no nudging) and increases to 1 (full nudging) for the highest model levels. Therefore large-scale circulation is fully constrained but near-surface winds are allowed to adapt to the surface land conditions as well as to the large-scale atmospheric conditions.

      Best! Kathrin


  • CC2: Comment on EGU2020-13802, Clemens Schwingshackl, 05 May 2020

    This is a really nice study, Kathrin!
    I have a question regarding the scaling plots on slide 5. For Northern Europe the CESM run agrees well with the other CMIP5 models but when you apply nudging, TX strongly increases. In the other regions, this effect is not so high and often CESM already lies in between the CMIP5 and the nudged runs. Does this mean, that in NEU the dynamics is the main (and maybe only?) contribution to the 2018 and similar events?

    • AC2: Reply to CC2, Kathrin Wehrli, 05 May 2020

      Thank you very much Clemens!

      Yes, I would say so! July TX strongly increases given this specific atmospheric circulation pattern, whereas it increases moderately (about 1°C for 1°C of global warming) given a random circulation. I would say that it is similar for CNA and ENA. On the other hand, averaged over MED there is no contribution by the dynamics but the CESM model is generally a bit on the warm side compared to other CMIP5 models. A side note at last: CMIP5-CESM uses interactive ocean so this makes it a bit difficult to fully disentangle the global warming from the circulation effect in my results.

Presentation version 1 – uploaded on 04 May 2020 , no comments