Poster presented at the ESA Living Planet Symposium 2025 in Session B.04.02 Addressing multi-hazards: Compounding and Cascading Events through Earth Observation Abstract: Recent advances in the study of compound and cascading extreme events have highlighted the growing need for extensive amounts of data to support empirical and systematic investigations. Events involving multiple hazards often have complex spatio-temporal dynamics, and they require robust data sources to enable meaningful analysis and improve the capacity of modelling their impacts. Despite the overwhelming increase in the availability of high resolution Earth Observation (EO) data in recent years, sampling compound and cascading extreme events remains challenging. This limits the current number of analysis-ready datasets. The lack of such datasets restricts the scope of scientific studies and hinders efforts to generalize their findings. Here, we introduce the ARCEME cascading event database, which is currently under development and whose aim is to sample cascading droughts and extreme precipitation events globally that impact society and ecosystems. The events, occurring between 2016 and 2024, are sampled from different hydro-climatic and socio-economic settings around the world. A dual event screening approach is used to identify the cascading events. The first approach relies on a time series of climate variables to detect weather extremes. The second approach uses records from the emergency database (EM-DAT), which resulted in human lives losses and economic damage. The overall result is a collection of cascading drought and extreme precipitation events that impacted society and/or on which to study their interaction with vegetation, leveraging multi-sensor EO data. The database will ultimately comprise data cubes with a spatial coverage of 10 by 10 km, of Sentinel-1 and Sentinel-2 images spanning 1 year before and one year after the extreme precipitation events. This structured approach ensures that the data captures the pre-event conditions, the event itself, and its aftermath, providing a comprehensive basis for detailed analysis. The ARCEME cascading event database aims at making available a high-quality, well structured analysis-ready database suitable for a wide range of applications like synoptic global studies on cascading and compound events, regional and local case studies on single extreme events or machine learning applications aiming at predicting impacts of extreme events on the vegetation state.