Study of Land Use Change in the Urmia Lake Water Shed Based on Landsat-TM Images and Pixel-Based and Object-Based Classification Techniques

Land use is one of the important features of watersheds, so being aware of its situation and changes over time are needed for achieving sustainable development. Today, novel techniques based on remote sensing technology provides good opportunity for study on this topic. In this study, the current and past status of land use of one of the most critical areas of country in regards of water and environmental issues, Lake Urmia watershed, was targeted. To this aim, Landsat -TM images from the years 1987 and 2007 were used and for improving the accuracy of classification, two sets of images from different months were used for each year. First, land use map of the Urmia County in 2007 was prepared by using of pixel-based (maximum likelihood) and object-based (fuzzy) methods, which Object-based approach showed significant more accuracy in classification of land use. Then object-based method was used to classify land use of whole basin for the two selected time periods.Theresultsindicate32and23percent growth in irrigated and rain fed cultivated area, respectively, as well as 8 percent reduction of rangelands in these two time periods. Also, surveying of land use change at involved provinces showed that the patterns of agricultural development were not identical among them.