Burned area assessment using Sentinel-2A satellite imagery and DNBR spectral index (Case study: forest areas on the anticline in the Khaiz region, Iran)

Abstract

The main advantage of incorporating remote sensing techniques into wildfire management is their ability to provide real-time data. This study aimed to investigate the extent of forest fires in southwestern Iran using remote sensing data. Sentinel-2A data with a resolution of 20 meters were used to conduct this study. It is worth noting that the spectral bands selected in this study, namely spectral band 8A (red edge 4) and band 12 (SWIR 2), have proved their suitability for fire intensity classification. In this study, 1NBR (Normalized Burn Ratio) values within the study area ranged from −0.096 to 0.81. These values were categorized based on the United States Geological Survey classification table. The study area covered 4,758.915 hectares, with approximately 32.41% (1,542.284 hectares) having calculated 1NBR values. Of the total area, 60.97% (2,901.675 hectares) was burned at low intensity, while approximately 6.62% (314.956 hectares) was burned at medium intensity. Unfortunately, due to the limited extent of the study area, regions with moderate to high fire intensity and high intensity were not included in the classification. The research results indicate that the studied index has satisfactory efficiency. The application of this index to regions with characteristics similar to those of the Khaiz anticline is likely to provide valuable and reliable results.