Monitoring and prediction of land use changes along expressways based on random forest and CA-Markov models: a case study of the Rongwu Expressway

To address the challenge of accurately capturing small-scale land use changes induced by highway construction, this study integrates the Random Forest algorithm with the CA-Markov model to conduct a fine-scale monitoring and simulation analysis. Focusing on the reconstruction and expansion section of the Rongwu Expressway (Yantai-Penglai segment) in Shandong Province, we utilized high-resolution Sentinel-2 remote sensing imagery to classify land use changes from 2020 to 2025. Furthermore, three scenarios—Natural Development, Cropland Protection Priority, and Ecological Protection Priority—were established to simulate and analyze future development trends for 2030. The results indicate that from 2020 to 2025, cropland decreased by 11.03%, while built-up land and forest land increased by 5.66% and 23.62%. In the multi-scenario simulations for 2030, distinct spatial trade-offs were observed among different development models; notably, the Cropland Protection Priority scenario proved most effective in curbing the expansion of built-up land and significantly increasing cropland retention. This study accurately delineates the spatial-temporal dynamics of land use induced by expressway expansion and reconstruction. The findings offer valuable insights for fine-grained territorial spatial governance and the mitigation of land-use conflicts along transportation infrastructure.