ANALISA PERBANDINGAN BERDASARKAN IDENTIFIKASI AREA KEBAKARAN DENGAN MENGGUNAKAN CITRA LANDSAT-8 DAN CITRA MODIS (Studi Kasus : Kawasan Gunung Bromo)

Abstract

Land and forest fire is a condition where the land and forest are hit by flame, causing land and forest destruction, economic losses, and environmental value. In the dry season, some of the regions in Indonesia often hit by flame. One example is a fire that occurred in the area of Mountain Bromo on October, 20th to 23rd 2014. These fires resulted in approximately 1.487 Ha of savanna forest got burned. Burned area can be identified by using remote sensing data, such as Landsat-8 and MODIS imagery. The method of identification used is the NDVI (Normalized Difference Vegetation Index) that utilizes near-infrared band and red band. Burned area is classified based on the value of the three threshold models. The use of multiple threshold models is done with the purpose of seeking the greatest accuracy value. Based on the results of burned area identification, obtained the greatest accuracy values of Landsat-8 is 48.394% from the μ - 1σ models and the wide of burned area is 1.354,5 Ha. While the greatest accuracy values of MODIS is 57,089% from the μ models and the wide of burned area is 1.005,209 Ha.