ACRS2015_5.pdf

ABSTRACT: The aim of this study is to conduct a study on land cover changes in Central part of Mongolian territory using multitemporal optical remote sensing (RS) data sets. As the data sources, Landsat images acquired in 1994, 2000, 2010 and 2014 are selected. To define the available land cover classes, the satellite images are classified using supervised (minimum distance, maximum likelihood and Mahalanobis distance) and unsupervised (isodata) classification methods and the results are compared. Of these methods, the better results are obtained through the use of the maximum likelihood method. Overall, the research indicates that in recent years, the land cover classes in the Central Mongolia have been changed.