Abstract
Three different data sets of images used to obtain the land cover changes in this study: Multi-Spectral Scanner (MSS) acquired in 1973, Landsat-5 Thematic Mapper (TM) acquired in 1990 and Landsat-8 Operational Land Imager (OLI) acquired 2013 consequently. For each data set, three Landsat scenes mosaicked to cover the whole study area. Supervised classification implemented to classify the area into six major land cover classes using two different classification algorithms. A total number of 400 points evenly distributed over the designated study area used in classification accuracy assessment. Kappa statistics obtained to specify the most appropriate classification algorithms in term of accuracy assessments. The results indicated that the rapid imbalance changes occurred among three land's cover classes urban area, surrounding desert and sedimentation. Human impacts in the form of sedimentation process practiced constantly on the Eastern Coast of Saudi Arabia, besides the loss of vegetation cover over the last four decades.