🔬 Researches on NRBR Index¶

📄 Normalized Radar Burn Ratio: A Case Study for Burned Area Mapping in Mediterranean Forests¶

Summary
This research introduces the Normalized Radar Burn Ratio (NRBR), an index designed to enhance burned area detection using Sentinel-1 C-band radar imagery. The research utilizes post- to pre-fire ratios of VV and VH backscatter coefficient to compute the NRBR, optimizing thus the contrast between burned and unburned areas. The 2017 wildfires in Portugal were used to validate the methodology. Using the U-Net architecture the NRBR based model outperforms previous ratio-based indices in metrics such as overall accuracy, omission error and Intersection over Union, among other metrics. Additionally, high correlations (r> 0.7) between NRBR and the optical indices NDVI (post-fire) and dNBR were observed. This approach has promising implications for improving burned area mapping, particularly for periods with cloud cover or occlusion from fire smoke.

Highlights
- 🌍 Region: Meditarranean forest
- 🛰️ Data: Sentinel-1 VV/VH polarizations
- 🤖 Model: U-Net, 512×512 patches
- 📈 Accuracy: Omission Error = 19% (vs. 32–66% for other indices)
- 🔍 Insight: Strong correlation with NDVI and dNBR (r > 0.7)

Citation
Tarazona, Y., Tanase, M.A., & Mantas, V. (2025). Normalized Radar Burn Ratio: A Case Study for Burned Area Mapping in Mediterranean Forests, IEEE Geoscience and Remote Sensing Letters., 2025. https://doi.org/10.1109/LGRS.2025.3592093

📄 Combining Ascending and Descending Sentinel-1 modes for Enhancing burned area mapping with Normalized Radar Burn Ratio¶

Summary
This study evaluates the Normalized Radar Burn Ratio (NRBR) for burned area mapping using Sentinel-1 SAR data, comparing ascending, descending, and combined acquisition modes of Sentinel-1. Results show that integrating both orbits significantly improves accuracy (0.921), recall (0.858), and reduces omission errors (14.2%), making NRBR a reliable tool for wildfire monitoring.

Highlights
- 🌍 Region: Meditarranean forest
- 🛰️ Data: Sentinel-1 VV/VH polarizations - 🤖 Model: U-Net, 512×512 patches
- 📈 Accuracy: Omission Error = 14% (vs. 18-19% for other indices)

Citation
Tarazona, Y., & Mantas, V. (2025). Combining Ascending and Descending Sentinel-1 modes for Enhancing burned area mapping with Normalized Radar Burn Ratio, Presented at Living Planet Symposium 2025. Zenodo. https://doi.org/10.5281/zenodo.15766613.

📄 Advancing Burned Area Mapping using the Normalized Radar Burn Ratio (NRBR)¶

🚧 Work in Progress¶

(Coming Soon!)

📄 Scalable Burned Area Mapping via Sentinel-1/2 Fusion using the new NRBR Index¶

🚧 Work in Progress¶

(Coming Soon!)