Evaluation spectral indicators to estimate burnt areas: Case study of Gholama Mountain, Tanomah area

Abstract

The purpose of this paper is to identify the burnt areas occurred in Gholama Mountain during the summer of 2020 using change detection analysis techniques. Multispectral images were obtained by Landsat 8. Satellite images of the same area at different times were compared, where landscape change were evaluated. In this paper, change detection technique was applied; to determine the burnt area and analyze it using several remote sensing indicators, in particular NBR index (Natural Burn Ratio) and the NDVI Index (Normalized Difference Vegetation Index) according to the case study in Gholama Mountain in Tanomah, using images derived from Landsat 8 images. The burnt areas were identified in a specific period and performance of the two indicators was evaluated. In fact, after creating confusion matrix for the two tested indicators, using methods, which indicate objective map quality (user accuracy, product accuracy, general accuracy and kappa coefficient), percentage values for each analyzed indicator were compared, and different methods for measuring accuracy revealed high quality for the results which can be achieved by the NBR index.