Veedu et al. (2025) Flood Forecasting Unveiled: Harnessing the Power of Sentinel-1A Imagery and ESA World Cover Through Multi-data Integration

Identification

• Journal: Lecture notes in networks and systems
• Year: 2025
• Date: 2025-10-25
• Authors: Jayasree Thazhath Veedu, Rajesh Reghunadhan
• DOI: 10.1007/978-3-032-02949-2_41

Research Groups

Department of Computer Science, Central University of Kerala, Kerala, India

Short Summary

This study investigates the integration of Sentinel-1A Synthetic Aperture Radar (SAR) imagery and ESA World Cover maps with deep neural networks to improve flood prediction accuracy, achieving 85.79% accuracy by combining these multi-source data.

Objective

• To enhance flood prediction accuracy by integrating multi-source remote sensing data, specifically Sentinel-1A Synthetic Aperture Radar (SAR) imagery (VV and VH channels) and the ESA World Cover map, using a deep neural network for semantic segmentation.

Study Configuration

• Spatial Scale: Regional/Local (implied by the use of high-resolution satellite imagery for flood mapping).
• Temporal Scale: Event-based (focus on flood events, utilizing recent satellite imagery).

Methodology and Data

• Models used: Deep neural network (for semantic segmentation).
• Data sources: Sentinel-1A Synthetic Aperture Radar (SAR) imagery (VV and VH polarization channels), ESA World Cover map.

Main Results

• A flood prediction accuracy of 85.79% was obtained by integrating VV and VH SAR channels with ESA World Cover map features.
• This combined approach yielded significantly better results compared to using only VV and VH SAR bands.
• The integration of multiple data sources (SAR imagery and ESA World Cover map) substantially enhanced the reliability of flood prediction models.

Contributions

• Demonstrates a significant improvement in flood prediction accuracy by combining Sentinel-1A SAR imagery with the ESA World Cover map.
• Highlights the value of multi-data integration for more robust and reliable flood prediction models.
• Contributes to better disaster preparedness and mitigation strategies through enhanced flood mapping capabilities.

Funding

• No specific funding information was provided in the paper text.

Citation

@article{Veedu2025Flood,
  author = {Veedu, Jayasree Thazhath and Reghunadhan, Rajesh},
  title = {Flood Forecasting Unveiled: Harnessing the Power of Sentinel-1A Imagery and ESA World Cover Through Multi-data Integration},
  journal = {Lecture notes in networks and systems},
  year = {2025},
  doi = {10.1007/978-3-032-02949-2_41},
  url = {https://doi.org/10.1007/978-3-032-02949-2_41}
}