Predicting land use change around railway stations: An enhanced CA-Markov model

Abstract

Predicting land use change around railway stations is crucial for facilitating the coordinated development of transport and land use. Previous studies have seldom focused on simulating and predicting small-scale land use changes at a railway station. Therefore, this study employs an enhanced Cellular Automata-Markov (CA-Markov) model, aiming to achieve simulations and predictions with heightened precision. Firstly, two additional driving factors, namely accessibility to railway stations and the kernel density of points of interest (POIs), are incorporated into the CA-Markov model. Secondly, the validation of the enhanced model is achieved through simulating land use changes around Dujiangyan Station. Finally, this model is applied to predict land use around Mianzhu South Station in 2026, and optimization strategies for railway station areas are proposed. The results indicate an 83.43-hectare reduction in farmland area, accompanied by a moderate increase in 35.8 hectares of forest land and 41.52 hectares of residential land. This enhanced model provides valuable technical support for strategic planning in railway station areas.