Nature Reviews Physics: Machine learning in fluid dynamics and climate physics

Learn more Watch now Add to Calendar 10/05/2022 04:00 PM 10/05/2022 05:10 PM Europe/London Nature Reviews Physics: Machine learning in fluid dynamics and climate physics Location of the event
Wednesday 05 Oct 2022
Time: 16:00 - 17:10

Event type

Virtual seminar

Audience type



Researchers in field of fluid dynamics have been experimenting with machine learning since the 1990s, having driven many advances in the use of these methods in modelling and simulation. The combination of real and simulated data together with physics-informed machine learning is now used in climate modelling. The extensive experience in benchmarking and validating fluid dynamics simulations can inform climate modelling and other fields.   

Watch now 

Chair - Dr. Gabriel Weymouth 

About the event

In this event, we will hear from Dr. Steven Brunton and Professor Laure Zanna. 

Dr. Steven Brunton - Machine learning for scientific discovery with examples in fluid mechanics 

I will describe how machine learning may be used to develop accurate and efficient nonlinear dynamical systems models for complex natural and engineered systems. I explore the sparse identification of nonlinear dynamics (SINDy) algorithm, which identifies a minimal dynamical system model that balances model complexity with accuracy, avoiding overfitting. This approach tends to promote models that are interpretable and generalisable, capturing the essential “physics” of the system. Additionally, I discuss the importance of learning effective coordinate systems in which the dynamics may be expected to be sparse. This sparse modeling approach will be demonstrated on a range of challenging modeling problems in fluid dynamics and I will discuss how to incorporate these models into existing model-based control efforts. As fluid dynamics is central to transportation, health and defence systems, I will emphasise the importance of machine learning solutions that are interpretable, explainable, generalisable and that respect known physics.

Professor Laure Zanna - Leveraging interpretable machine learning for climate physics 

In this presentation, I will describe the complex and multiscale nature of the climate system and how machine learning can be leveraged to deepen our understanding of key physical climate processes. I will focus on advances in interpretable and physics-aware machine learning methods that have the potential to accelerate scientific discovery in climate physics and modeling. In particular, I will discuss examples of interpretable and generalisable machine learning models that capture ocean turbulence processes (horizontal scale of 10 km-100 km) and how these turbulent features can impact large-scale ocean currents (1000’s of kms). The machine-learned models of turbulent processes are shown to improve coarse-resolution climate simulations by faithfully capturing the complex multiscale dynamical properties in the climate system. 



Professor Ben MacArthur

Director of AI for Science and Government, Deputy Programme Director for Health and Medical Sciences, and Turing Fellow

Professor Jonathan Rowe

AI for Science and Government Programme Chair, Programme Director for Data Science for Science, and Turing Fellow