NVDI- and NBR-based Comparison for Burnt Areas Characterization Using Sentinel-2 Imagery

Mapping the risk of areas susceptible to forest fires requires the use of new techniques such as remote sensing and geographic information systems (GIS), which are complementary tools to the conventional methods used to undertake prevention and response management actions. Accurate and rapid mapping of fire-affected areas is fundamental to support fire management, take account of environmental losses, define planning strategies and monitor vegetation restoration. Remote sensing is used for forest fire detection, monitoring and post-fire vegetation regrowth. Remote sensing data can be used to locate active fires and subsequent hotspots. Comparison between burnt areas and active fires provides information on fire speed and direction. The integration of satellite imagery in the fight against forest fires will make it possible to identify areas affected by forest fires, particularly those that are inaccessible, and to improve forest fire assessments by providing a more refined and accurate delimitation, and by taking into account burnt areas in inaccessible zones that Forest managers were unable to estimate. Annual monitoring of forest fires using satellite imagery gives a global view and provides very useful information on the environment, for a wide range of scales, and establishes a status of forest areas covered by fire, mainly during heatwave periods. To characterize burnt areas, we can use the Normalized Difference Vegetation Index (NDVI) or Normalized Burn Ratio (NBR). This study tries to measure the difference between these two methodologies. In results, the NBR index gives results that are more precise, it only indicates burnt surfaces areas, whereas the NDVI can include bare land surfaces for example.