Strong El Niño episodes are characterized by warm sea surface temperatures and strong thunderstorm activity in the eastern tropical Pacific, but their influence is nearly global. Such episodes can induce severe droughts and wildfires in some faraway regions and intense rainfall and catastrophic floods in others. These remote linkages, or drivers, are examples of atmospheric teleconnections, and they play a critical role in regional climate variability.
Often, the physics and dynamics underlying such teleconnections are neither well understood nor accurately represented by state-of-the-art climate models. Overcoming this knowledge gap will be key to improving not only future regional climate change projections but also seasonal forecasts. Improving projections and forecasts would have huge societal benefits, especially in regions that are particularly vulnerable to climate change impacts. To better understand atmospheric teleconnections, the next generation of climate scientists must be holistically trained in the underlying physics, climate modeling, and innovative data analyses techniques.
One project, called Globally Observed Teleconnections and Their Role and Representation in Hierarchies of Atmospheric Models (GOTHAM), aims to better understand large-scale modes of climate variability and their mutual interactions. This project is funded by the Belmont Forum and the Joint Programming Initiative “Connecting Climate Knowledge for Europe” (JPI Climate). GOTHAM combines novel complex system-based analysis methods with superensembles of atmospheric models generated through distributed computing using the citizen science platform climateprediction.net (also known as CPDN).
Within the framework of this project, 27 aspiring young scientists and several experts on teleconnection patterns and data science attended the GOTHAM International Summer School in Potsdam, Germany. Similar to GOTHAM itself, the school program integrated diverse topics, including midlatitude flow, stratospheric teleconnections, the El Niño–Southern Oscillation, and monsoons. The program also focused on the interplay between these different teleconnections and the affected regions. (A detailed report of the summer school can be found here.)
In addition to in-depth lectures and student poster sessions, practical exercises provided hands-on training for CPDN and several Python-based complex systems tools. For example, one group of students used CPDN data on the unique atmospheric circulation pattern of summer 2010, which induced severe weather extremes like the heat wave in Russia and floods in Pakistan. Advanced, complex network-based analysis techniques were applied to a large ensemble of simulations of that summer to help understand the underlying physical pathways.
Our experience with this summer school demonstrates that integral training on both theory and practice can be achieved within even the short period of only 1 week. The participating young researchers obtained important foundational training for applying their newly gathered methodological skills to their specific research questions. They also started new collaborations with climate and data experts.
Finally, the school underscored the necessity and importance of combining expertise from different disciplines: climate scientists, data scientists, and seasonal forecast experts. Integrating these different fields constitutes a crucial step toward better understanding the physics behind and the role of atmospheric teleconnections.
The authors thank Scott Osprey and all GOTHAM partners for their support in organizing this school. Financial support has been provided by the German Federal Ministry for Education and Research (grants 01LP1611A and 01LN1306A).
—Eftychia Rousi (email: [email protected]), Potsdam Institute for Climate Impact Research, Germany; Dim Coumou, Potsdam Institute for Climate Impact Research, Germany; also at Institute for Environmental Studies, Vrije Universiteit Amsterdam, Netherlands; and Reik V. Donner, Potsdam Institute for Climate Impact Research, Germany