MetNet: A Neural Weather Model for Precipitation Forecasting

TL;DR

MetNet is a neural network model that predicts high-resolution, short-term precipitation maps using radar and satellite data, outperforming traditional numerical weather prediction for up to 8 hours ahead.

Contribution

Introduces MetNet, a novel neural network architecture with axial self-attention for high-resolution, probabilistic weather forecasting at unprecedented spatial and temporal scales.

Findings

MetNet outperforms Numerical Weather Prediction up to 8 hours ahead.

Achieves 1 km² spatial resolution with 2-minute temporal updates.

Provides probabilistic precipitation forecasts with high accuracy.

Abstract

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.