Math + Explainable AI for weather and climate, with a focus on extremes

Add to your list(s) Download to your calendar using vCal

• Gianmarco Mengaldo -- NUS, Singapore
• Wednesday 12 June 2024, 12:00-13:00
• Wolfson Lecture Theatre.

If you have a question about this talk, please contact Adriano Gualandi.

Earth’s climate is changing rapidly under the effect of global warming, leading to more frequent and severe extreme weather events [1,2]. These weather extremes, in turn, are exacting heavy socioeconomic and environmental tolls [3], prompting an urgent need for better understanding and predicting them. In this talk, we present some recent results obtained for the tropical Indo-Pacific region, using human-understandable methods (or taking a human view), namely dynamical system theory. In particular, we show that changes in weather patterns are leading to more weather extremes, namely heatwaves and extreme precipitation. We then present the use of explainable AI tools (i.e., machine view) to investigate the onset and precursors of these extremes. More specifically, we try to bridge existing human knowledge (human view) and “AI knowledge” (machine view) to better understand the behaviour and predictability of weather extremes.

This talk is part of the Bullard Laboratories Wednesday Seminars series.

This talk is included in these lists:

• Bullard Laboratories Wednesday Seminars
• Department of Earth Sciences seminars
• Wolfson Lecture Theatre

Note that ex-directory lists are not shown.