On Tuesday, July 2, we organized a working session of the NL AIC Water and Climate Working Group, in which we explored with representatives from the government, market and knowledge communities how we can make a solid contribution to solving climate change issues with the help of AI.
Summary
The session started with two inspiring presentations by Roberto Bentivoglio (TU-Delft) and Hans van Beek (TU Eindhoven) on the insights there are from science on the use of AI in extreme situations (inside the outside the water domain). The sheets, presented toolboxes and links to more information can be found at the bottom of this page. The participants concluded after the presentations that water is steering for many social and economic functions, underlining the importance of being able to make better predictions around extreme weather.
In the second part of the meeting, participants worked in groups on concrete use cases. This led to the following use cases:
- Making satellite data accessible to end users in water management: AI for VARAA (Visualization, Accelleration, Representativeness, Accuracy, Actuality)
- The digital dream of AI emulators: a key technology for water and climate. Develop a model alternative that is good enough (almost as predictive as the original, perhaps also more robust) but above all very much faster than the original.
- AI for data preprocessing of large datasets (standardization, metadata, etc).
- AI for intervention selection in disaster situations.
In the coming period, we will start to highlight these cases one by one through interviews on our website. The cases will be further developed in several follow-up meetings.
More information
- Werkgroep Water en Klimaat NL AIC
- Robert Bentivoglio: Paper ‘Rapid spatio-temporal flood modelling via hydraulics-based graph neural networks’ with links to the corresponding code and dataset
- Hans van Beek: all information and software regarding Bayesian Inference can be found on the BIASlab website
