Forest Fire Area DetectionUsing Earth Observation Satellites: A Case Study of Ain Marah Al-Jabal Al-Akhdar, Libya

Abstract

A forest fire started on June 30th, 2021and was stopped on July 3rd  in Ain Marah Al Jabal Al-Akhdar causing damage and casualties. There had not been any official results of the area burnt by the fire. This study is an attempt to provide an accurate burnt space area using Normalized difference vegetation(NDVI) index,  Burned Area Index (BAI), Normalized Burn Ratio (NBR) indices, which was derived from the European satellite Sentinel-2A. the indices were calculated for the displacements of burned and unburned pixels in the pre-/post-fire NIR-MIR and NIR-R bi-spectral spaces. NDVI classified the space of actual vegetation loss by 96.8 ha, while NBR and BAI were calculated for the same images in order to gain the actual burned area which were estimated by 103 and 101.8 ha respectively. The estimated burned area obtained from NBR and BAI were slightly vary that is because the area had already passed through an unusual long hot weather before the fire occurrence, making the part vegetation cover already dry and look as  bare land. Additional the land cover is overlapped by low dry  vegetations and bare land, that may disturb the calculations.