The influence of landscape heterogeneity on the dynamic urban growth simulation based on cellular automata model: The case of Wuhan metropolitan area, China

This study proposed a LH method based on the landscape pattern model, and constructed a cellular automata (CA) prediction model of urban growth considering the LH. The results showed that we integrated the binary logistic regression (BLR)-CA and particle swarm optimization (PSO)-CA model to simulate the urban growth for different periods in the Wuhan metropolitan area. It was found that the simulation results generated by the CA model considering LH are significantly improved, indicating that historical LH information has a significant impact on the possibility and speed of future urban growth. Furthermore, the LH not only affects the urban growth, but also causes the urban growth to produce the redistribution of LH, emphasizing that the importance of considering the LH of transition rules in the simulation of urban growth. The results have important practical significance for urban growth modeling, planning, and management.