Land use change detection in Denizli City Center using spectral angle mapper method and evaluations in terms of some earth science data

Abstract

Land use planning is central for a sustainable development processes and should be structured in harmony with geo-environmental constrains. In this study, land use change detection was determined in Denizli residential area between the years 1984 and 2018. In the analyses, Landsat-5TM and Landsat-8 OLI images were used for 1984 and 2018, respectively. The geometric, radiometric calibration and atmospheric corrections were applied to the satellite images in preprocessing stage. Land use change detection was performed using supervised classification -Spectral Angle Mapper- method. Accuracy of the classifications was evaluated by the kappa index, which was 0.80 for 1984 and 0.87 for 2018 image. According to the results, between 1984 and 2018 there was an increase of 155% and 96% in built up and forest areas, while 52% and 40% decrease were observed in agriculture and bare land, respectively. It was also determined that the majority of the settlement areas increased in regions close to active fault zones. Accordingly, the built up areas within the 500 m buffer zone to the active faults were increased by 240% from 1984 to 2018.