Monitoring and Spatially Explicit Simulation of Land Use Dynamics: From Cellular Automata to Geosimulation - A Case Study of Tehran, Iran

In this investigation, several simulation approaches such as cellular automata, the Markov chain model, CA-Markov model, hybrid logistic regression model and agent-based modeling (ABM) are employed in order to simulate urban expansion. The complexity of the aforementioned models increase respectively once the available environmental and socio-economic variables are taken into consideration. However, the comparison of the models results in developing a customized ABM, which takes the individual behaviors into consideration. Within this ABM, three different agents influencing urban sprawl are defined: developer agents, resident agents, government agents. Each agent considers the relevant variables, which are controlled by the respective agent. In addition, the importance of each variable, which affects urban sprawl, was identified through an analytic hierarchy process analysis. Each particular model was evaluated statistically in order to assure the validity of the approach and afterwards, the approach was applied to predict the forthcoming urban expansion.