The research aim is application of GIS and spatial analysis of satellite images for agricultural mapping. The study area is located in Mecsek Hills, Hungary. These landscapes represent unique part of Hungarian environment in the Carpathian Basin. Diverse landscape structure with complex biogeographic composition characterizes this region. However, this region is intensely used in agriculture, which causes high land heterogeneity. The research method consists in using ILWIS GIS and Landsat TM satellite imagery (1992, 1999 and 2006). The methodology includes clustering, spatial analysis, interpretation and classification. Clustering algorithm is based on the analysis of the similarity of the spectral signatures of pixels, and grouping pixels into thematic categories of classes of the land cover types in study area. This paper contributes to the development of the agricultural monitoring in region of central Europe. The distribution of the landscapes in the Mecsek Hills area was analyzed. Land cover types were interpreted by association of pixels on the image to different thematic classes: vegetation, categories of agricultural land and other areas (in total, 12 classes). The results consists in thematic land cover maps of the study area in 1992, 1999, 2006 showing landscape dynamics.