Deep Learning for Precipitation Nowcasting: A Benchmark and A New Model

Abstract

With the goal of making high-resolution forecasts of regional rainfall, precipitation nowcasting has become an important and fundamental technology underlying various public services ranging from rainstorm warnings to flight safety. Recently, the Convolutional LSTM (ConvLSTM) model has been shown to outperform traditional optical flow based methods for precipitation nowcasting, suggesting that deep learning models have a huge potential for solving the problem. However, the convolutional recurrence structure in ConvLSTM-based models is location-invariant while natural motion and transformation (e.g., rotation) are location-variant in general. Furthermore, since deep-learning-based precipitation nowcasting is a newly emerging area, clear evaluation protocols have not yet been established. To address these problems, we propose both a new model and a benchmark for precipitation nowcasting. Specifically, we go beyond ConvLSTM and propose the Trajectory GRU (TrajGRU) model that can actively learn the location-variant structure for recurrent connections. Besides, we provide a benchmark that includes a real-world large-scale dataset from the Hong Kong Observatory, a new training loss, and a comprehensive evaluation protocol to facilitate future research and gauge the state of the art.

NIPS 2017 Spotlight

References in corpus (4)

• Generating Videos with Scene Dynamics
• Decomposing Motion and Content for Natural Video Sequence Prediction
• Video (language) modeling: a baseline for generative models of natural videos
• Unsupervised Learning for Physical Interaction through Video Prediction

Cited by in corpus (26)

• A trans-disciplinary review of deep learning research for water resources scientists
• Skillful Precipitation Nowcasting using Deep Generative Models of Radar
• WeatherBench: A benchmark dataset for data-driven weather forecasting
• A Review on Deep Learning Techniques for Video Prediction
• GaAN: Gated Attention Networks for Learning on Large and Spatiotemporal Graphs
• Deep Learning for Post-Processing Ensemble Weather Forecasts
• A Novel Framework for Spatio-Temporal Prediction of Environmental Data Using Deep Learning
• SwinVRNN: A Data-Driven Ensemble Forecasting Model via Learned Distribution Perturbation
• Statistical post-processing of wind speed forecasts using convolutional neural networks
• Effective Training Strategies for Deep-learning-based Precipitation Nowcasting and Estimation
• Nowcasting-Nets: Deep Neural Network Structures for Precipitation Nowcasting Using IMERG
• Hybrid physics-AI outperforms numerical weather prediction for extreme precipitation nowcasting
• A Deep Learning Method for Real-time Bias Correction of Wind Field Forecasts in the Western North Pacific
• Accurate and Diverse Sampling of Sequences based on a "Best of Many" Sample Objective
• RNDiff: Rainfall nowcasting with Condition Diffusion Model
• CloudCast: A Satellite-Based Dataset and Baseline for Forecasting Clouds
• Deep learning for improved global precipitation in numerical weather prediction systems
• Accurate and Clear Precipitation Nowcasting with Consecutive Attention and Rain-map Discrimination
• Routing algorithms as tools for integrating social distancing with emergency evacuation
• Reduced-Gate Convolutional LSTM Using Predictive Coding for Spatiotemporal Prediction
• Interpretable Crowd Flow Prediction with Spatial-Temporal Self-Attention
• Travel Speed Prediction with a Hierarchical Convolutional Neural Network and Long Short-Term Memory Model Framework
• Dynamic-structured Semantic Propagation Network
• Deep Vision in Analysis and Recognition of Radar Data: Achievements, Advancements and Challenges
• A Multi-task Two-stream Spatiotemporal Convolutional Neural Network for Convective Storm Nowcasting
• Spatio-temporal Video Re-localization by Warp LSTM