Urban Expansion Modeling with Logistic Regression

Today, due to the limited natural resources of land, rapid population growth, rapid expansion of cities, future land use prediction is very important for land managers, planners and environmental specialists because land use change effect on ecosystem and also threaten vital resources. Modeling and analysis of the phenomenon of urban development provide comprehensive information to urban planners and managers to have better and more scientific planning. The main objective of this research is modeling urban growth for the city of Sanandaj, in the west of Iran using satellite imagery, Geographic Information Systems and logistic regression. The parameters are used in this study, including distance to developed area, distance to main roads, distance to green spaces, elevation, slope, distance to fault, distance to district centers and number of urban cell in a 3 by 3 neighborhood. Figure of Merit, Kappa coefficient and Percent Correct Match (PCM) have been used to evaluate goodness of fit of proposed model.