Agent-based modeling and simulations of land-use and land-cover change according to ant colony optimization: a case study of the Erhai Lake Basin, China

The land-use structure and ecological service functions of the Erhai Lake Watershed are being altered by rapid socioeconomic development and urbanization, which will ultimately lead to the generation and aggravation of agricultural and urban non-point source pollution over the entire region. Therefore, the relationships between human activities and land-use/land-cover changes (LUCCs) must be studied to support scientific decisions regarding reasonable land planning and land use. This paper combines geographic information system technology for spatial analysis and the ant colony optimization artificial intelligence algorithm. Moreover, this study applies agent-based modeling to establish a spatiotemporal process model for LUCCs that effectively simulates the dynamic land-use changes in the basin. A selection is first made and evaluated for dynamic land-use change impact factors. Then, the agent classes and their rules in the LUCC processes are established. The program is designed using the Java programming language, and the model is implemented based on the Repast modeling platform. Finally, the models are validated, and the simulated results are analyzed and discussed. Some conclusions were drawn from the experiments, as well as some policies on land use were suggested.