Global‐Frequency Synergy: A Novel Paradigm for Radar Echo Extrapolation via Attention and Fourier Convolution

Wei Wu, Guangxin He, Xiaoran Zhuang, Yuxuan Feng, Juanzhen Sun, Haonan Chen, Lei Lei, Jingjia Luo

Journal of Geophysical Research Atmospheres · 23 May 2026 · Colorado State University (incl. CIRA), NSF NCAR / UCAR, Nanjing University of Information Science & Technology (NUIST)

Paper

In one sentence. GFST-LSTM adds attention and Fourier convolution to ST-LSTM for radar echo extrapolation, improving CSI at 40 dBZ by 22.9% over PredRNN.

Abstract

Abstract Accurate radar echo extrapolation is critical for short‐term weather forecasting, yet existing deep learning methods often suffer from echo ambiguity, intensity decay, and insufficient global context utilization. To address these limitations, this paper proposes Global‐Frequency Spatiotemporal Long Short‐Term Memory ( GFST‐LSTM), a novel model that integrates a global attention mechanism and Fourier convolutional modules into the Spatiotemporal LSTM ( ST‐LSTM) architecture. The attention module dynamically weights multi‐scale spatiotemporal features by enhancing channel and spatial correlations, while the Fourier convolution module captures global periodic patterns via frequency‐domain transformations. Evaluated on the Moving Modified National Institute of Standards and Technology database (Moving MNIST) benchmark and Jiangsu Province radar data sets (2019–2021), GFST‐LSTM achieves a 22.9% improvement in Critical Success Index and 13.1% in Heidke Skill Score over Predictive Recurrent Neural Network at the 40 dBZ threshold. Notably, it excels in preserving strong echo regions during 60–120 min predictions, reducing positional bias by 6.6% compared to the Motion Gated Recurrent Unit (MotionGRU). Ablation studies confirm the synergistic effect of both modules, with the full model outperforming variants that lack either component.

Labels

Labels and summary are generated by Claude and reviewed weekly. Wrong? Edit this entry.