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

Rossby wave breaking through the 21st century in a global climate model

Kevin Bowley1 and Melissa Gervais1,2
Kevin Bowley and Melissa Gervais
  • 1Penn State University, Department of Meteorology and Atmospheric Science, University Park, United States of America (
  • 2Penn State University, Institute for Computational and Data Sciences, University Park, United States of America (

Rossby wave breaking on the dynamic tropopause (DT) occurs when synoptic-scale Rossby waves become highly amplified and undergo a breaking process.  This process can result in significant meridional transport of air masses resulting and intrusions of low latitude air poleward, high latitude air equatorward, or a combination of the two.  The ensuing modification of the troposphere and lower stratosphere in response to such events have been areas of considerable research due to their potential impacts on both high- and low-frequency mid- and high-latitude variability.  Furthermore, the processes and feedbacks associated with these events can result in notable changes to the jet structure and are frequently associated with atmospheric river events amongst other phenomena.  As such, the potential impacts of future changes in these events make them of considerable interest for identifying and studying in global climate model (GCM) simulations. 

Here, we apply a Rossby wave breaking identification scheme to three sets of 25-member Community Earth System Model simulations with prescribed sea surface temperature and sea ice conditions over the historical period (2010-2019), mid-Century (2050-2059) and late-Century (2090-2099).  This dataset represents a unique opportunity to study Rossby wave breaking processes in future climate simulations on a dynamically evolving surface rather than the more common pressure levels or isentropic levels as the DT is calculated for each of the CESM members.  Both anticyclonic and cyclonic Rossby wave breaking events are identified and tracked.  Events modeled in the historical period are compared to existing reanalysis data for the same period to explore the ability of the CESM model in this configuration to reproduce these events accurately.  Furthermore, the three periods of interest are examined to determine changes in the locations of Rossby wave breaking as well as the dynamic and thermodynamic characteristics of composited events. 

How to cite: Bowley, K. and Gervais, M.: Rossby wave breaking through the 21st century in a global climate model, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-19980,, 2020


Display file

Comments on the display

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

displays version 3 – uploaded on 24 Apr 2020
Upload error on previous versions.
  • CC1: Comment on EGU2020-19980, Olivia Martius, 27 Apr 2020

    Hi Kevin 

    Very interesting results! Do you know about potential links of the changes in wavebreaking to blocking? Do you see similar zonal shifts and changes in frequency in blocking? 



    • AC1: Reply to CC1, Kevin Bowley, 29 Apr 2020

      Good evening Olivia - I'm glad you find the work interesting!  The linkages between blocking and wave breaking is something we've been certainly thinking about quite a bit for these future projections.  In general, we find that the CESM under-represents wave breaking in a similar manner as is seen for blocking.  We've postulated that the challenges in resolving these features are essentially one in the same, and that the under-represenations in the CESM are primarily due to challenges in the model accurately resolving topography, SSTs, and transient eddies.  Regarding the downstream shift, our results show some similarity to those shown in Woolings et al. 2018 (Current Climate Change Reports (2018) 4:287–300 that nicely summarized the challenges of representing blocking as well as the potential for downstream shifts.  Their figure 4 shows this downstream shift in blocking using the anomaly method for identifying blocking which has many similarities to what we see with our wave breaking signal.  As such, I think we have some confidence in suggesting that these two features are coupled, but we haven't explored that avenue yet.  The sensitivity to the blocking method (and, I could imagine, the wave break identification method, though we haven't explored this) to the locations of positive and negative trends in these features does raise an interesting question about the most effective methodology toward addressing this question that certainly should be explored in the future.  Thanks so much!


displays version 2 – uploaded on 24 Apr 2020, no comments
Small editorial change to motivation slide.
displays version 1 – uploaded on 23 Apr 2020, no comments