Achieving Conservation of Energy in Neural Network Emulators for Climate Modeling

2019/06/15 by Beucler, Tom, Rasp, Stephan, Pritchard, Michael +1 · 1 citation
#Atmospheric and Oceanic Physics (physics.ao-ph) #Computational Physics (physics.comp-ph) #FOS: Computer and information sciences #FOS: Physical sciences #Machine Learning (cs.LG)

paper · doi:10.48550/arxiv.1906.06622

Abstract

Artificial neural-networks have the potential to emulate cloud processes with higher accuracy than the semi-empirical emulators currently used in climate models. However, neural-network models do not intrinsically conserve energy and mass, which is an obstacle to using them for long-term climate predictions. Here, we propose two methods to enforce linear conservation laws in neural-network emulators of physical models: Constraining (1) the loss function or (2) the architecture of the network itself. Applied to the emulation of explicitly-resolved cloud processes in a prototype multi-scale climate model, we show that architecture constraints can enforce conservation laws to satisfactory numerical precision, while all constraints help the neural-network better generalize to conditions outside of its training set, such as global warming.

Cited by

• Physically consistent model learning for reaction-diffusion systems
• Diffeomorphic Neural Operator Learning

Related

• Towards Physically-consistent, Data-driven Models of Convection
• Deep learning to represent subgrid processes in climate models
• Exploring Randomly Wired Neural Networks for Climate Model Emulation
• Can Machines Learn How Clouds Work? The Epistemic Implications of Machine Learning Methods in Climate Science
• Stable Emulation of an Entire Suite of Model Physics in a State-of-the-Art GCM using a Neural Network