MetNet: A Neural Weather Model for Precipitation Forecasting
This addresses weather forecasting, a critical problem with social and economic impacts, by providing a novel deep learning approach that improves short-term predictions.
The paper tackles precipitation forecasting by introducing MetNet, a neural weather model that predicts precipitation up to 8 hours ahead at high resolution, outperforming Numerical Weather Prediction for forecasts up to 7-8 hours on a continental scale.
Weather forecasting is a long standing scientific challenge with direct social and economic impact. The task is suitable for deep neural networks due to vast amounts of continuously collected data and a rich spatial and temporal structure that presents long range dependencies. We introduce MetNet, a neural network that forecasts precipitation up to 8 hours into the future at the high spatial resolution of 1 km$^2$ and at the temporal resolution of 2 minutes with a latency in the order of seconds. MetNet takes as input radar and satellite data and forecast lead time and produces a probabilistic precipitation map. The architecture uses axial self-attention to aggregate the global context from a large input patch corresponding to a million square kilometers. We evaluate the performance of MetNet at various precipitation thresholds and find that MetNet outperforms Numerical Weather Prediction at forecasts of up to 7 to 8 hours on the scale of the continental United States.