Non-linear and heterogeneous relationship between proximity to high-speed rail stations and land value in China: Analysis using XGBoost-SHAP modelling

High-speed rail (HSR) stations serve as pivotal nodes in transportation networks and are widely recognized as catalysts in shaping surrounding land use. This study investigates the relationship between proximity to HSR stations and land value in China from 2008 to 2022, using XGBoost-SHAP modeling. The findings highlight that proximity to HSR stations plays a significant role in predicting land value, exhibiting a non-linear relationship: land values are higher for land parcels within a specific distance threshold to HSR stations, although this effect has weakened over time. The influence on predicting land value is particularly pronounced for special-class and intra-city HSR stations in mega-cities, while proximity thresholds vary depending on the station grade, city type, and spatial context. Notably, proximity to HSR stations contributes more to commercial land value predictions in mega-cities and very large cities, and to industrial land value predictions in small-to-medium and large cities. These insights provide valuable guidance for policymakers and urban planners seeking to optimize land development strategies around HSR stations, and offers tailored recommendations to address diverse regional and contextual needs.