NeuralGCM combines traditional atmospheric physics with machine learning to simulate the climate more quickly, without compromising on accuracy.
NeuralGCM is a hybrid climate model developed by Google Research, which combines classical physical equations of atmospheric circulation with machine-learning components. This approach aims to reconcile the physical rigour of traditional climate models with the computational speed enabled by AI.
The stakes are high: comprehensive climate models usually require considerable computational resources; an AI-accelerated version could make it possible to test a much greater number of scenarios in order to refine regional climate projections.
Read the full article on the original source
Go to the source →