AI-based Image & Text Analysis and Citizen Sensing to Improve Warnings for Extreme Weather Events and their Impacts

Purpose and goal

This project aims to assess the potential of combining AI-based image processing and text mining with national impact-based weather warning systems. Integrating expertise from national and regional agencies on climate adaptation and weather warning systems, climate science and policy research, visualization and AI, the proposed project explores if and to what extent AI-based algorithms can be employed to evaluate the accuracy of impact-based weather warnings, and assesses the added value of integrating AI-based information into the existing weather warning systems.

Expected results and effects

This project aims to explore the capacity of AI-based image processing and text mining to contribute to evaluating the accuracy of the national system for impact-based weather warnings and to potentially contribute to the further development of the system in order to increase resilience to extreme weather events, which are expected to be more frequent due to climate change.

Planned approach and implementation

This project sets out for three main tasks: (1) Assessment of approaches for collecting image and text data appropriate for AI-based analysis derived from citizen science campaigns and social media (2) Development of machine learning algorithms for text and image analysis (3) Development and assessment of the application and results as part of a co-design process with climate adaptation experts, as well as with experts for the SMHI national weather warning system and involved authorities at local, regional and central level