Google Research· ResearchModel update

NeuralGCM harnesses AI to better simulate long-range global precipitation

Google Research describes an update to NeuralGCM, a hybrid atmospheric model that combines physics-based modeling with a neural network. The new version is trained directly on NASA satellite-based precipitation observations (2001–2018) instead of reanalysis data, leading to improved simulations of average precipitation, precipitation extremes (especially the top 0.1% of rainfall), and the daily weather cycle. At 280 km resolution, it outperforms a leading operational model from ECMWF on 2–15 day forecasts and shows improvements for multi-decadal climate simulations.

From the source

Now in “Neural general circulation models for modeling precipitation”, published in Science Advances, we describe how NeuralGCM was trained on satellite-based precipitation observations to achieve improved simulations of precipitation. Notably, at the current resolution of 280 km, we see improvements against a leading operational model for medium-range weather forecasting (up to 15 days) and against atmospheric models used for multi-decadal climate simulations. We find NeuralGCM more accurately reproduces average precipitation, precipitation extremes — with major improvements for the top 0.1% of rainfall — and the daily weather cycle.

research.google