Budapest, 1–4 April 2025. The OpenSense COST Action successfully held its latest Training School on Merging and Application of Opportunistic Rainfall Sensor Data at the Budapest University of Technology and Economics. The event brought together around 35 participants, including PhD candidates, master students, early-career researchers, and professionals from across Europe, all eager to deepen their understanding and skills in working with opportunistic rainfall sensors.

Practical, Hands-On Learning

Over the four-day program, participants engaged in intensive hands-on sessions and lectures, focusing on both fundamental methods and real-world applications of OS data, such as from Commercial Microwave Links (CML), Personal Weather Stations (PWS), and Smartphone-based Measurements (SML). The training was hosted in Room KMf26 (Oltay), Building K, offering a collaborative space for learning and exchange.

Each core topic was led by experienced trainers:

• Intro to working with OS data – Dr. Maximilian Graf (DWD)
• Merging of OS into rainfall products – Dr. Christian Chwala (KIT), Erlend Øydvin (NMBU)
• Application 1: OS-based nowcasting with pysteps – Dr. Ruben Imhoff (Deltares), Jenna Ritvanen (FMI)
• Application 2: Hydrological modelling with OS data – Dr. Martin Fencl (CTU Prague), Dr. Jochen Seidel (University of Stuttgart).

Spotlight on Mini-Projects

One of the highlights of the Training School was the Mini-Project component, where trainees formed small interdisciplinary teams to apply what they had learned in a creative and applied context. These mini-projects focused on practical case studies, data merging exercises, and exploratory applications of OS data using real datasets and open-source tools.

Topics explored by the teams included:

• Fusion of CML and PWS data for urban rainfall mapping
• Nowcasting with pysteps in regions with limited radar coverage
• Evaluation of merged OS datasets against reference observations
• Integrating OS data into hydrological runoff models
• Assessing the value of CML-based rainfall data in sparse gauge networks

The projects culminated in short presentations on the final day, where participants showcased their approaches, challenges, and initial results. These sessions not only demonstrated the potential of OS data for rainfall monitoring but also fostered lively discussion and collaboration among early-career researchers and experts.