Affiliation:
1. IZMIR KATIP CELEBI UNIVERSITY
Abstract
Forest fires are the leading natural disasters that endanger the living and lifeless environment in forest ecosystem. Every year, millions of forested areas burn out and Turkey is one of the countries most affected by the forest fires. In this study, forest fire started near Karabaglar was investigated. Using Landsat 8 satellite images, fire area was detected by using five different remote sensing indexing methods (dNDVI, dNBR, dNBRT, dBAI, RBR) were used and the fire area is mapped. Fire intensity was calculated and related map is then obtained. Thus, burn severity distribution in the region was estimated and shown as the burn severity map. As a result of the analysis, it is found that the most convenient results come from the RBR analyses with a 99% of consistency when they are compared with the burn severity data obtained in the field by the General Directorate of Forestry. The remote sensing technique has also demonstrated its ability to distinguish damage severity levels, even including undamaged sections in an entire forest fire damage zone, in a way that cannot be completely done in a field work.
Publisher
International Journal of Environment and Geoinformatics
Subject
General Arts and Humanities