Zhang et al. (2026) MS-EdgeCast: A Dual-Stage Framework With a Multiscale Convolutional Recurrent Network and Edge-Guided Diffusion for Convective Storm Nowcasting

Identification

• Journal: IEEE Transactions on Geoscience and Remote Sensing
• Year: 2026
• Date: 2026-01-01
• Authors: Wei Zhang, Lintao Wang, Muqi Luo, Renbo Pang, Xiaojiang Song, Xiangguang Zhang
• DOI: 10.1109/tgrs.2026.3655487

Research Groups

[N/A]

Short Summary

This paper introduces MS-EdgeCast, a dual-stage framework that integrates a multiscale convolutional recurrent network and edge-guided diffusion for the purpose of convective storm nowcasting.

Objective

• To develop and evaluate MS-EdgeCast, a novel dual-stage framework combining a multiscale convolutional recurrent network and edge-guided diffusion, for enhanced convective storm nowcasting.

Study Configuration

• Spatial Scale: [N/A, typically mesoscale to local scales for nowcasting]
• Temporal Scale: [N/A, typically minutes to a few hours for nowcasting]

Methodology and Data

• Models used: MS-EdgeCast (a dual-stage framework), Multiscale Convolutional Recurrent Network, Edge-Guided Diffusion
• Data sources: [N/A]

Main Results

[N/A]

Contributions

[N/A, likely improved accuracy or robustness in convective storm nowcasting through the proposed novel framework]

Funding

[N/A]

Citation

@article{Zhang2026MSEdgeCast,
  author = {Zhang, Wei and Wang, Lintao and Luo, Muqi and Pang, Renbo and Song, Xiaojiang and Zhang, Xiangguang},
  title = {MS-EdgeCast: A Dual-Stage Framework With a Multiscale Convolutional Recurrent Network and Edge-Guided Diffusion for Convective Storm Nowcasting},
  journal = {IEEE Transactions on Geoscience and Remote Sensing},
  year = {2026},
  doi = {10.1109/tgrs.2026.3655487},
  url = {https://doi.org/10.1109/tgrs.2026.3655487}
}