Land Cover Changes Based on Landsat Imagery Interpretation

This paper presents the use of satellite data (i.e., Landsat-5 & Landsat-8) to interpret change land cover from 1997 2020. The study area covers administrative boundary Lumajang Regency. land-cover map year derived Landsat-5. Land-cover 2020 interpreted Landsat-8. uses two methods image classifications unsupervised and supervised). procedure includes enhancement, registration, classification. Then, classification results evaluated by confusion-matrix (overall kappa accuracy). supervised produces 7 classes Land forest, pavement/urban area), paddy field, plantation, rural area, water body sand mining area. Unsupervised produced four 5 class i.e., built-area, plantation. Supervised done overall accuracy = 86% 82%, while 73% 64% for imagery. Furthermore, image, reaches 93% 90%, 81% 72%. method gives a better result than un-supervised. Comparison 2020, it also shows increase in pavement or build-area, followed sand-mining. appears as decrease forest plantation areas.Keywords: Landsat-5, Landsat-8, Unsupervised, Supervised,