Advancing burned area mapping using the Normalized Radar Burn Ratio (NRBR)

Earth Observation data provides valuable and comprehensive information for monitoring land cover disturbances, including wildfire impacts. This study focuses on the Mediterranean climate zone and assesses the effectiveness of the new Normalized Radar Burn Ratio (NRBR) index for burned area detection. Derived from Sentinel-1 C-Synthetic Aperture Radar (SAR) data using a U-net deep learning model, we provide the first regional-scale assessment of NRBR across diverse land cover types, demonstrating that radar-based NRBR can match the performance of the widely used optical-based delta Normalized Burn Ratio (dNBR). The analysis explores the sensitivity of NRBR to post-fire structural changes across different land cover types. Results show that NRBR achieved an Intersection over Union (IoU) of 64.8% and a Dice Coefficient (DC) of 79.3% in forested areas, while obtained slightly higher values (IoU: 85.8%, DC: 91.6%). In grassland covers, NRBR outperformed (IoU: 57.6% vs. 53.4%; DC: 73.0% vs. 69.5%), demonstrating a comparable capability for burned area detection relative to . These findings emphasize the potential of NRBR as a SAR-based index to complement optical approaches and improve the reliability of burned area mapping, particularly under adverse atmospheric conditions. Further validation across diverse ecosystems is recommended to consolidate the operational applicability of NRBR for burned area monitoring across the world.