Enter Note Done
Go to previous page in this tab
Session
  • Presentation | IN13B: Large Language Models and Agentic Workflows in Science: Applications, Safety, and Geoscience Innovations II Poster
  • Poster
  • Bookmark Icon
  • IN13B-0332: Transforming Climate Services with LLMs and Multi-Source Data Integration
  • Schedule
    Notes
  • Board 0332‚ Hall EFG (Poster Hall)
    NOLA CC
    Set Timezone

Generic 'disconnected' Message
Author(s):
Ivan Kuznetsov, Alfred Wegener Institute Helmholtz-Center for Polar and Marine Research Bremerhaven (First Author, Presenting Author)
Antonia Anna Jost, Alfred Wegener Institute Helmholtz-Center for Polar and Marine Research Bremerhaven
Dmitrii Pantiukhin, Alfred Wegener Institute Helmholtz-Center for Polar and Marine Research Bremerhaven
Boris Shapkin, Alfred Wegener Institute Helmholtz-Center for Polar and Marine Research Bremerhaven
Thomas Jung, Alfred Wegener Institute Helmholtz Centre for Polar and Marine Research
Nikolay Koldunov, Alfred Wegener Institute Helmholtz-Center for Polar and Marine Research Bremerhaven


ClimSight is an open-source digital assistant that turns complex climate information into clear, location-specific guidance. It works by combining trusted climate datasets with a “large language model” (LLM) - an advanced type of artificial intelligence that can read and write text much like a human. When someone asks a question such as “How will heat and rainfall affect wheat near Marrakech in the 2040s?”, ClimSight automatically pulls the best available climate projections, checks key thresholds for the crop, and explains the risks and possible adaptations in everyday language.


Tests on dozens of real-world questions show that adding local climate numbers and specialised data-retrieval “agents” helps the system give more complete and accurate answers than a generic chatbot. Because ClimSight can run on cost-efficient AI models, it offers a practical way for farmers, city planners, insurers and emergency managers to access the climate insights they need - without having to decode technical reports. This work demonstrates how pairing transparent AI with high-quality climate data can widen public access to climate services and support better decisions in a warming world.




Scientific Discipline
Suggested Itineraries
Neighborhood
Type
Main Session
Discussion