Land use change detection using post classification comparison LandSat satellite images (Case study: land of Tehran)

Knowing the ratio of landuse and how this ratio change over the time is a significant factor in landuse planning and management. By knowing how landuses change over time, we can predict future changes and do preventive and appropriate actions. The aim of this study is detecting the landuse changes in the Tehran city in period of (1995-2005). Landuse maps for the mentioned years were derived from LandSat imagery and improved using AutoCAD data produced by the National Cartographic Center (NCC) and available maps. For verification produced maps, using the error matrix approach, a Kappa coefficient of 88% was achieved and after classification, change detection was carried out using comparison approach. The results of change detection with a Kappa coefficient of 91% and Cramer’s V of 88% show that the greatest increases were in built up area (4603.68 ha), parks (650.7 ha) and water level (22.32 ha) and in contrast, the greatest decreases were in open land (4561.47 ha) and vegetation cover (715.23 ha). In addition, according to the results the most changes of built up area has been occurring in the west part of Tehran.