Artificial intelligence for climate impacts:
Applications in flood risk

Source: Nature Portfolio


Published on the website of the journal Nature, this contribution by researchers Anne Jones, Julian Kuehnert, Paolo Fraccaro, Ophélie Meuriot, Tatsuya Ishikawa, Blair Edwards, Nikola Stoyanov, Sekou L. Remy, Kommy Weldemariam et Solomon Assefa  have created an AI-based application, specifically Machine Learning, to quantify and model flood risks.

Figure 1 - Time series illustrating flood prediction uncertainty quantification

To develop their program, researchers Anne Jones, Julian Kuehnert, Paolo Fraccaro, Ophélie Meuriot, Tatsuya Ishikawa, Blair Edwards, Nikola Stoyanov, Sekou L. Remy, Kommy Weldemariam and Solomon Assefa used Bayesian optimization with Gaussian process regression to computational fluid dynamics problems.


Previous
Previous

AI and the Climate Emergency: Opportunities, Challenges, and Recommendations

Next
Next

Detection of Bow Echoes in Kilometer-Scale Forecasts Using a Convolutional Neural Network