MetNet: A Neural Weather Model for Precipitation Forecasting

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.

References in corpus (4)

• Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting
• Pixel Recurrent Neural Networks
• Large Scale Adversarial Representation Learning
• Axial Attention in Multidimensional Transformers

Cited by in corpus (24)

• Skillful Precipitation Nowcasting using Deep Generative Models of Radar
• A Generative Deep Learning Approach to Stochastic Downscaling of Precipitation Forecasts
• VideoGPT: Video Generation using VQ-VAE and Transformers
• Short-term precipitation prediction using deep learning
• Revisiting the Calibration of Modern Neural Networks
• A review of radar-based nowcasting of precipitation and applicable machine learning techniques
• Effective Training Strategies for Deep-learning-based Precipitation Nowcasting and Estimation
• Dual-Branched Spatio-temporal Fusion Network for Multi-horizon Tropical Cyclone Track Forecast
• Global Extreme Heat Forecasting Using Neural Weather Models
• Cloud Cover Nowcasting with Deep Learning
• TempEE: Temporal-Spatial Parallel Transformer for Radar Echo Extrapolation Beyond Auto-Regression
• MS-nowcasting: Operational Precipitation Nowcasting with Convolutional LSTMs at Microsoft Weather
• Accurate and Clear Precipitation Nowcasting with Consecutive Attention and Rain-map Discrimination
• Convolutional LSTM Neural Networks for Modeling Wildland Fire Dynamics
• Generative Models as Distributions of Functions
• Deep multi-stations weather forecasting: explainable recurrent convolutional neural networks
• PlumeNet: Large-Scale Air Quality Forecasting Using A Convolutional LSTM Network
• Variational Data Assimilation with a Learned Inverse Observation Operator
• S3RP: Self-Supervised Super-Resolution and Prediction for Advection-Diffusion Process
• Skillful Twelve Hour Precipitation Forecasts using Large Context Neural Networks
• STAA: Spatio-Temporal Alignment Attention for Short-Term Precipitation Forecasting
• Modeling Wildfire Perimeter Evolution using Deep Neural Networks
• Generation and Simulation of Yeast Microscopy Imagery with Deep Learning
• PlumeCityNet: Multi-Resolution Air Quality Forecasting