RAILJUST · High-Speed Rail, Environmental Inequality, and Spatial Variation in China: A Causal Machine Learning Approach

Infrastructure development can foster economic growth but may also exacerbate environmental inequalities by creating uneven exposure to air pollution. This project examines the environmental impacts of China’s high-speed rail network, with a focus on the particulate matter with a diameter of 2.5 micrometers or less (PM2.5) pollution and its unequal distribution across regions. Using satellite-derived PM2.5 data and causal, interpretable machine learning methods, the project will (i) estimate the impact of high-speed rail on pollution inequality and (ii) uncover spatial variation in its drivers. By integrating remote sensing, causal inference, and machine learning, the research advances interdisciplinary approaches to assessing infrastructure impacts and environmental justice. The results will provide insights relevant not only to China but also to Europe and beyond, supporting policy debates on sustainable infrastructure, regional cohesion, and the EU Green Deal.