Co-Creating Narrative-Based Near-Term Climate Predictions to Prepare for Volatile Future Climates (part of Excellence Cluster CLICCS / Phase II)

The project brings together advances in physical sciences, artificial intelligence (AI) in climate modelling, and qualitative research methods. Its goal is to develop comprehensible narratives for near-term climate information within both social and natural sciences. This involves exploring "narrative modes" for representing scientific knowledge under conditions of uncertainty, extending beyond individual cognition. The project also combines conceptual and predictive system development to evaluate how these approaches influence the realization of desired climate futures. One key aspect is redesigning prediction systems to improve their accuracy (prediction skill) by integrating different physics-based narratives. These systems are designed to be easily customizable to meet diverse societal interests and needs. To achieve this, the project employs ensemble selection techniques, recent AI advancements, and comprehensive visual analysis methods. These tools contribute to creating redesigned systems that facilitate a systematic application of physical storytelling in climate predictions. The project will also test these systems at real-world sites of knowledge co-creation across CLICCS II project. The aim is to examine how scientific knowledge—particularly in the form of near-term climate narratives—is perceived, negotiated, and contested across different regions, political regimes, and levels of climate vulnerability.