Quantification of Burned Severity of the Forest Fire using Sentinel-2 Remote Sensing Images: A Case Study in the Ella Sri Lanka

Abstract

Forest fire far could be considered as one of the majors concerning environmental issues mainly in tropical climate regions. In Sri Lanka, forest plantations and “sparsely used croplands” are the further most vulnerable areas of a forest fire. The case study was based on the forest fire reported in the Ella Rock region in 2019. The remote sensing techniques were utilized for the analysis in the QGIS open-source environment through Semi-automatic Classification Plugin (SCP) and Sentinel-2 images employed as the key source of data. Normalized Burn Ratio (NBR) and Normalized Difference Vegetation Index (NDVI) were initially applied on the pre and post images and then computed the difference of NBR (dNBR) and the difference of NDVI (dNDVI). Then through the change detection techniques extent of the fire and the severity levels were obtained. As revealed by the investigation 73.82 hectares of areas were burned due to the forest fire and 15.65% of the area was highlighted as a high severity of the burn. Moreover, NDVI and NBR significantly important in forest fire mapping also emphasized by the study. The unavailability of a complete database of the forest fire in Sri Lanka found as the major issue. Further, taking necessary actions to prevent forest fire a vital requirement of the current context.

Cite as

Sandamali K.U.J., & Chathuranga K.A.M. (2021). Quantification of Burned Severity of the Forest Fire using Sentinel-2 Remote Sensing Images: A Case Study in the Ella Sri Lanka. Research and Reviews: Journal of Environmental Sciences, 3(3), 1–12. http://doi.org/10.5281/zenodo.5115527