Lin et al. (2026) Self-Supervised Temporal Super-Resolution Framework with METEOR-Net for Enhancing Convective Process Reconstruction in Geostationary Satellite Cloud Data

Identification

• Journal: IEEE Transactions on Geoscience and Remote Sensing
• Year: 2026
• Date: 2026-01-01
• Authors: Xiang Lin, Yunying Li, Zitong Chen, Xiongwei Kou
• DOI: 10.1109/tgrs.2026.3677111

Research Groups

[Information not available in the provided text.]

Short Summary

This paper introduces METEOR-Net, a self-supervised temporal super-resolution framework designed to enhance the reconstruction of rapidly evolving convective processes using geostationary satellite cloud data.

Objective

• To develop and apply a self-supervised temporal super-resolution framework (METEOR-Net) to improve the temporal resolution and reconstruction accuracy of convective processes observed in geostationary satellite cloud data.

Study Configuration

• Spatial Scale: Geostationary satellite coverage, focusing on cloud data, likely at typical satellite pixel resolutions (e.g., 0.5 km to 4 km).
• Temporal Scale: Enhancing temporal resolution from a lower frequency (e.g., hourly) to a higher frequency (e.g., 10-15 minutes) to better capture the rapid evolution of convective processes.

Methodology and Data

• Models used: METEOR-Net (a self-supervised temporal super-resolution framework, likely based on deep learning architectures such as convolutional neural networks).
• Data sources: Geostationary satellite cloud data (observation data).

Main Results

• [Quantitative results are not available in the provided text.]
• The framework is expected to significantly improve the temporal resolution of geostationary satellite cloud observations.
• It likely leads to a more accurate and detailed reconstruction of the initiation, development, and dissipation phases of convective systems.
• The self-supervised nature potentially reduces the need for extensive paired high-resolution ground truth data.

Contributions

• Introduction of a novel self-supervised temporal super-resolution framework (METEOR-Net) specifically tailored for geostationary satellite cloud data.
• Advancement in the capability to monitor and reconstruct rapidly evolving convective processes, which are crucial for nowcasting and severe weather prediction.
• Offers a method to enhance the utility of existing geostationary satellite data without requiring additional high-frequency observations for training.

Funding

[Information not available in the provided text.]

Citation

@article{Lin2026SelfSupervised,
  author = {Lin, Xiang and Li, Yunying and Chen, Zitong and Kou, Xiongwei},
  title = {Self-Supervised Temporal Super-Resolution Framework with METEOR-Net for Enhancing Convective Process Reconstruction in Geostationary Satellite Cloud Data},
  journal = {IEEE Transactions on Geoscience and Remote Sensing},
  year = {2026},
  doi = {10.1109/tgrs.2026.3677111},
  url = {https://doi.org/10.1109/tgrs.2026.3677111}
}