EXTREMES at EGU2025: Presenting research on weather and climate extremes in Serbia

The EXTREMES project team participated in the European Geosciences Union (EGU) General Assembly 2025, held in Vienna, Austria, from April 27 to May 2, 2025. As one of the largest international gatherings of geoscientists, EGU provides an invaluable platform for presenting cutting-edge research, exchanging ideas, and building interdisciplinary collaborations.

In the months leading up to the conference, our team worked intensively to prepare high-quality contributions. From finalizing data analyses to refining visual content and coordinating abstract submissions, every step was a team effort aimed at presenting impactful science aligned with the broader goals of the EXTREMES project.

At EGU25, we contributed to several sessions covering a wide range of topics including weather and climate extremes, compound events, climate change impacts, and climate services. Our team members shared findings through poster presentations and also contributed as co-conveners of an innovative short course.

Highlights of Our Contributions:

🔬 Submicron-size aerosol scavenging and ice nucleation

TM2, TM3, TM9 Link to abstract The cloud physics team presented recent work on electrostatic scavenging of atmospheric particles by cloud droplets and its influence on ice nucleation processes. This is particularly relevant for the Central Balkans—Europe's second most favorable region for thunderstorm development—where hail poses significant socio-economic risks. The study offers new insights into how operational cloud seeding practices in Serbia might be influenced by electroscavenging mechanisms, adding a layer of understanding to ongoing hail suppression efforts.

🌩️ Convective indices and severe hail in Serbia

TM3, TM2 Link to abstract This contribution explored the behavior and trends of convective indices during days with severe hail events in Serbia. Understanding these patterns is essential for enhancing early warning systems.

❄️ Climatology of freezing rain in Belgrade (1949–2022)

TM5, TM4, TM6 Link to abstract Presented in poster format, this study examined the long-term characteristics of freezing rain in Serbia’s capital. Key findings show that freezing rain events—mostly occurring during winter—are associated with specific synoptic conditions and have decreased in frequency over the past decades, likely influenced by climate change. This type of localized climate insight is crucial for urban planning and infrastructure resilience.

🤖 Short course on Deep Learning for extreme events

TM7, TM8 In collaboration with international colleagues, TM7 and TM8 co-convened a popular short course titled "Detecting and Characterizing Extreme Events with Deep Learning". The session introduced xtlim, a new open-source Python library designed for identifying and analyzing extreme events from large and complex datasets. The course engaged a diverse audience and highlighted the role of AI and deep learning in advancing climate risk assessments. The integration of xtlim into the EXTREMES project toolbox will enhance our capacity to analyze extreme event indices across Serbia and the broader region.

Looking Ahead

EGU25 was not only an opportunity to present our research, but also to gain valuable feedback, form new collaborations, and stay at the forefront of climate science innovation. We are excited to integrate these insights into our ongoing work and continue strengthening the role of science in informing climate adaptation strategies across Southeastern Europe.

Stay tuned for more updates from the EXTREMES project as we continue pushing the boundaries of climate research and communication!