Integration of Remote Sensing data and GIS for prediction of land cover map Falahatkar Samereh1,*, Soffianian Ali Reza1, Khajeddin Sayed Jamaleddin1, Ziaee Hamid Reza2, Nadoushan Mozhgan Ahmadi1 1Natural resource faculty, Iafahan University of technology, Isfahan, Iran 2Municipality of Isfahan, Iran *Email: s7falahatkar@yahoo.com
Online published on 3 January, 2012. Abstract Satellite remote sensing and geographic information system (GIS) have been widely applied in identifying and analyzing land use and land cover change. In this study, aerial photos and MSS, TM and ETM+ images of Isfahan and its environs were used to provide maps of land cover for 1955, 1972, 1990 and 2001. A hybrid method was used for image classification: a combination of supervised and unsupervised classification. Cellular automata filter facilities and Markov model were used together in a CAMarkov model for predicting land cover maps. To study the accuracy of the predicted maps and validation of the CAMarkov model, three methods were used: a calculating agreement and disagreement table, a chisquared goodnessoffit test and an error matrix. The results indicate that if land cover change processes are constant, the CA Markov model predicts land cover changes for the following years, for which agreement between predicted maps with actual map is less than 70% in this study area. Top Keywords Change detection, CAMarkov model, Land cover. Top |