Climate change is affecting extreme weather – we identify how and when society is at risk

Seasonal to sub-seasonal (S2S) predictability could provide societies with valuable information on weather-related risk, allowing decision-makers to initiate early warning action plans and to optimize resource management. Using Artificial Intelligence (AI), we are developing a rigorous data-driven framework that enables automatic detection of interpretable physical drivers on S2S timescales. Combining expert knowledge, causal inference and a variety of machine learning techniques, our tool has worked for science. Now we aim for societal impact by making predictions for stakeholders (e.g. NGO’s).